pgd_adversarial_training.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "f3789cea",
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "import torch.nn.functional as F\n",
    "import torch.backends.cudnn as cudnn\n",
    "\n",
    "import torchvision\n",
    "import torchvision.transforms as transforms\n",
    "\n",
    "import os\n",
    "import time\n",
    "from models import  LinfPGDAttack, ResNet18"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "3ba6cd06",
   "metadata": {},
   "outputs": [],
   "source": [
    "learning_rate = 0.1\n",
    "epsilon = 0.0314\n",
    "k = 7\n",
    "alpha = 0.00784\n",
    "file_name = 'pgd_adversarial_training'\n",
    "\n",
    "device = 'cuda' if torch.cuda.is_available() else 'cpu'\n",
    "\n",
    "transform_train = transforms.Compose([\n",
    "    transforms.RandomCrop(32, padding=4),\n",
    "    transforms.RandomHorizontalFlip(),\n",
    "    transforms.ToTensor(),\n",
    "])\n",
    "\n",
    "transform_test = transforms.Compose([\n",
    "    transforms.ToTensor(),\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "328a2361",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Files already downloaded and verified\n",
      "Files already downloaded and verified\n"
     ]
    }
   ],
   "source": [
    "train_dataset = torchvision.datasets.CIFAR10(root='./data', train=True, download=True, transform=transform_train)\n",
    "test_dataset = torchvision.datasets.CIFAR10(root='./data', train=False, download=True, transform=transform_test)\n",
    "\n",
    "train_loader = torch.utils.data.DataLoader(train_dataset, batch_size=128, shuffle=True, num_workers=4)\n",
    "test_loader = torch.utils.data.DataLoader(test_dataset, batch_size=100, shuffle=False, num_workers=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "e9e10cc0",
   "metadata": {},
   "outputs": [],
   "source": [
    "net = ResNet18()\n",
    "net = net.to(device)\n",
    "net = torch.nn.DataParallel(net)\n",
    "cudnn.benchmark = True\n",
    "\n",
    "adversary = LinfPGDAttack(net,epsilon,alpha,k)\n",
    "criterion = nn.CrossEntropyLoss()\n",
    "optimizer = optim.SGD(net.parameters(), lr=learning_rate, momentum=0.9, weight_decay=0.0002)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "90aa64d4",
   "metadata": {},
   "outputs": [],
   "source": [
    "def train(epoch):\n",
    "    print('\\n[ Train epoch: %d ]' % epoch)\n",
    "    net.train()\n",
    "    train_loss = 0\n",
    "    correct = 0\n",
    "    total = 0\n",
    "    for batch_idx, (inputs, targets) in enumerate(train_loader):\n",
    "        inputs, targets = inputs.to(device), targets.to(device)\n",
    "        optimizer.zero_grad()\n",
    "\n",
    "        adv = adversary.perturb(inputs, targets)\n",
    "        adv_outputs = net(adv)\n",
    "        loss = criterion(adv_outputs, targets)\n",
    "        loss.backward()\n",
    "\n",
    "        optimizer.step()\n",
    "        train_loss += loss.item()\n",
    "        _, predicted = adv_outputs.max(1)\n",
    "\n",
    "        total += targets.size(0)\n",
    "        correct += predicted.eq(targets).sum().item()\n",
    "        \n",
    "        if batch_idx % 10 == 0:\n",
    "            print('\\nCurrent batch:', str(batch_idx))\n",
    "            print('Current adversarial train accuracy:', str(predicted.eq(targets).sum().item() / targets.size(0)))\n",
    "            print('Current adversarial train loss:', loss.item())\n",
    "\n",
    "    print('\\nTotal adversarial train accuarcy:', 100. * correct / total)\n",
    "    print('Total adversarial train loss:', train_loss)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "f3f65d08",
   "metadata": {},
   "outputs": [],
   "source": [
    "def test(epoch):\n",
    "    print('\\n[ Test epoch: %d ]' % epoch)\n",
    "    net.eval()\n",
    "    benign_loss = 0\n",
    "    adv_loss = 0\n",
    "    benign_correct = 0\n",
    "    adv_correct = 0\n",
    "    total = 0\n",
    "    with torch.no_grad():\n",
    "        for batch_idx, (inputs, targets) in enumerate(test_loader):\n",
    "            inputs, targets = inputs.to(device), targets.to(device)\n",
    "            total += targets.size(0)\n",
    "\n",
    "            outputs = net(inputs)\n",
    "            loss = criterion(outputs, targets)\n",
    "            benign_loss += loss.item()\n",
    "\n",
    "            _, predicted = outputs.max(1)\n",
    "            benign_correct += predicted.eq(targets).sum().item()\n",
    "\n",
    "            if batch_idx % 10 == 0:\n",
    "                print('\\nCurrent batch:', str(batch_idx))\n",
    "                print('Current benign test accuracy:', str(predicted.eq(targets).sum().item() / targets.size(0)))\n",
    "                print('Current benign test loss:', loss.item())\n",
    "\n",
    "            adv = adversary.perturb(inputs, targets)\n",
    "            adv_outputs = net(adv)\n",
    "            loss = criterion(adv_outputs, targets)\n",
    "            adv_loss += loss.item()\n",
    "\n",
    "            _, predicted = adv_outputs.max(1)\n",
    "            adv_correct += predicted.eq(targets).sum().item()\n",
    "\n",
    "            if batch_idx % 10 == 0:\n",
    "                print('Current adversarial test accuracy:', str(predicted.eq(targets).sum().item() / targets.size(0)))\n",
    "                print('Current adversarial test loss:', loss.item())\n",
    "\n",
    "    print('\\nTotal benign test accuarcy:', 100. * benign_correct / total)\n",
    "    print('Total adversarial test Accuarcy:', 100. * adv_correct / total)\n",
    "    print('Total benign test loss:', benign_loss)\n",
    "    print('Total adversarial test loss:', adv_loss)\n",
    "\n",
    "    state = {\n",
    "        'net': net.state_dict()\n",
    "    }\n",
    "    if not os.path.isdir('checkpoint'):\n",
    "        os.mkdir('checkpoint')\n",
    "    torch.save(state, './checkpoint/' + file_name)\n",
    "    print('Model Saved!')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "518d325d",
   "metadata": {},
   "outputs": [],
   "source": [
    "def adjust_learning_rate(optimizer, epoch):\n",
    "    lr = learning_rate\n",
    "    if epoch >= 100:\n",
    "        lr /= 10\n",
    "    if epoch >= 150:\n",
    "        lr /= 10\n",
    "    for param_group in optimizer.param_groups:\n",
    "        param_group['lr'] = lr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "b922330c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "[ Train epoch: 0 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.0\n",
      "Current adversarial train loss: 4.878740310668945\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.09375\n",
      "Current adversarial train loss: 2.9484665393829346\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.1328125\n",
      "Current adversarial train loss: 2.475701332092285\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.1171875\n",
      "Current adversarial train loss: 2.486109495162964\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.15625\n",
      "Current adversarial train loss: 2.35545015335083\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.140625\n",
      "Current adversarial train loss: 2.336601495742798\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.1484375\n",
      "Current adversarial train loss: 2.305589199066162\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.1171875\n",
      "Current adversarial train loss: 2.341203451156616\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.1796875\n",
      "Current adversarial train loss: 2.239138126373291\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.1640625\n",
      "Current adversarial train loss: 2.2376208305358887\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.140625\n",
      "Current adversarial train loss: 2.2789828777313232\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.1640625\n",
      "Current adversarial train loss: 2.29023814201355\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.1484375\n",
      "Current adversarial train loss: 2.2839791774749756\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.125\n",
      "Current adversarial train loss: 2.2405598163604736\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.1953125\n",
      "Current adversarial train loss: 2.192243814468384\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.171875\n",
      "Current adversarial train loss: 2.197737693786621\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.1953125\n",
      "Current adversarial train loss: 2.164329767227173\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.2578125\n",
      "Current adversarial train loss: 2.125203847885132\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.140625\n",
      "Current adversarial train loss: 2.2718474864959717\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.15625\n",
      "Current adversarial train loss: 2.2748355865478516\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.2109375\n",
      "Current adversarial train loss: 2.0946462154388428\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.2109375\n",
      "Current adversarial train loss: 2.173060417175293\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.2109375\n",
      "Current adversarial train loss: 2.1692049503326416\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.1328125\n",
      "Current adversarial train loss: 2.198371171951294\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.1796875\n",
      "Current adversarial train loss: 2.1305532455444336\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.1875\n",
      "Current adversarial train loss: 2.1602070331573486\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.203125\n",
      "Current adversarial train loss: 2.128568172454834\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.1796875\n",
      "Current adversarial train loss: 2.169095277786255\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.203125\n",
      "Current adversarial train loss: 2.122964859008789\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.234375\n",
      "Current adversarial train loss: 2.141005516052246\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.21875\n",
      "Current adversarial train loss: 2.0520966053009033\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.2109375\n",
      "Current adversarial train loss: 2.1121582984924316\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.2109375\n",
      "Current adversarial train loss: 2.1293466091156006\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.234375\n",
      "Current adversarial train loss: 2.0472395420074463\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.1875\n",
      "Current adversarial train loss: 2.1588902473449707\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.2421875\n",
      "Current adversarial train loss: 2.049396276473999\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.15625\n",
      "Current adversarial train loss: 2.1735990047454834\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.1953125\n",
      "Current adversarial train loss: 2.105928421020508\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.203125\n",
      "Current adversarial train loss: 2.054716110229492\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.2375\n",
      "Current adversarial train loss: 2.079749345779419\n",
      "\n",
      "Total adversarial train accuarcy: 17.53\n",
      "Total adversarial train loss: 889.7007548809052\n",
      "\n",
      "[ Test epoch: 0 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.4\n",
      "Current benign test loss: 1.7676246166229248\n",
      "Current adversarial test accuracy: 0.34\n",
      "Current adversarial test loss: 1.9178612232208252\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.36\n",
      "Current benign test loss: 1.9552377462387085\n",
      "Current adversarial test accuracy: 0.3\n",
      "Current adversarial test loss: 2.1155433654785156\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.3\n",
      "Current benign test loss: 1.881402850151062\n",
      "Current adversarial test accuracy: 0.22\n",
      "Current adversarial test loss: 2.0342555046081543\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.39\n",
      "Current benign test loss: 1.8492944240570068\n",
      "Current adversarial test accuracy: 0.3\n",
      "Current adversarial test loss: 2.0154268741607666\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.18\n",
      "Current benign test loss: 1.9948276281356812\n",
      "Current adversarial test accuracy: 0.14\n",
      "Current adversarial test loss: 2.164984941482544\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.29\n",
      "Current benign test loss: 1.9257862567901611\n",
      "Current adversarial test accuracy: 0.22\n",
      "Current adversarial test loss: 2.113776922225952\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.27\n",
      "Current benign test loss: 1.892136812210083\n",
      "Current adversarial test accuracy: 0.19\n",
      "Current adversarial test loss: 2.0445656776428223\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.3\n",
      "Current benign test loss: 1.9267500638961792\n",
      "Current adversarial test accuracy: 0.22\n",
      "Current adversarial test loss: 2.1078107357025146\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.36\n",
      "Current benign test loss: 1.8477635383605957\n",
      "Current adversarial test accuracy: 0.28\n",
      "Current adversarial test loss: 2.0059096813201904\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.23\n",
      "Current benign test loss: 1.9058668613433838\n",
      "Current adversarial test accuracy: 0.17\n",
      "Current adversarial test loss: 2.070370674133301\n",
      "\n",
      "Total benign test accuarcy: 31.5\n",
      "Total adversarial test Accuarcy: 24.36\n",
      "Total benign test loss: 188.75528383255005\n",
      "Total adversarial test loss: 204.83075559139252\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 1 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.265625\n",
      "Current adversarial train loss: 2.0076894760131836\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.2109375\n",
      "Current adversarial train loss: 2.164552927017212\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.1875\n",
      "Current adversarial train loss: 2.1142666339874268\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.2734375\n",
      "Current adversarial train loss: 2.0079870223999023\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.1953125\n",
      "Current adversarial train loss: 2.0869505405426025\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.265625\n",
      "Current adversarial train loss: 2.1380434036254883\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.2578125\n",
      "Current adversarial train loss: 2.0209217071533203\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.203125\n",
      "Current adversarial train loss: 2.1096580028533936\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.265625\n",
      "Current adversarial train loss: 2.065950393676758\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.2265625\n",
      "Current adversarial train loss: 2.0325567722320557\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.2265625\n",
      "Current adversarial train loss: 1.9853755235671997\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.2265625\n",
      "Current adversarial train loss: 1.9904712438583374\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.234375\n",
      "Current adversarial train loss: 2.0166537761688232\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.2109375\n",
      "Current adversarial train loss: 2.0726959705352783\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.21875\n",
      "Current adversarial train loss: 2.076143980026245\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.1875\n",
      "Current adversarial train loss: 2.018078565597534\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.28125\n",
      "Current adversarial train loss: 2.043539524078369\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.234375\n",
      "Current adversarial train loss: 2.0736942291259766\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.234375\n",
      "Current adversarial train loss: 2.0854599475860596\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.203125\n",
      "Current adversarial train loss: 2.021531581878662\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.25\n",
      "Current adversarial train loss: 1.986951470375061\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.21875\n",
      "Current adversarial train loss: 2.050636053085327\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.2109375\n",
      "Current adversarial train loss: 2.0911378860473633\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.2734375\n",
      "Current adversarial train loss: 2.0164906978607178\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.328125\n",
      "Current adversarial train loss: 1.9156787395477295\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.2734375\n",
      "Current adversarial train loss: 1.9739367961883545\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.25\n",
      "Current adversarial train loss: 2.0416243076324463\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.1953125\n",
      "Current adversarial train loss: 2.083329916000366\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.2734375\n",
      "Current adversarial train loss: 2.007772922515869\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.2578125\n",
      "Current adversarial train loss: 2.001054286956787\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.25\n",
      "Current adversarial train loss: 2.0564281940460205\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.234375\n",
      "Current adversarial train loss: 1.9687687158584595\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.234375\n",
      "Current adversarial train loss: 1.9767166376113892\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.2890625\n",
      "Current adversarial train loss: 2.0607333183288574\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.265625\n",
      "Current adversarial train loss: 1.9494493007659912\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.3359375\n",
      "Current adversarial train loss: 1.9778310060501099\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.2578125\n",
      "Current adversarial train loss: 2.019904851913452\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.2421875\n",
      "Current adversarial train loss: 1.9293766021728516\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.2578125\n",
      "Current adversarial train loss: 1.9609993696212769\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.1375\n",
      "Current adversarial train loss: 2.1617093086242676\n",
      "\n",
      "Total adversarial train accuarcy: 23.964\n",
      "Total adversarial train loss: 795.6733430624008\n",
      "\n",
      "[ Test epoch: 1 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.46\n",
      "Current benign test loss: 1.6282540559768677\n",
      "Current adversarial test accuracy: 0.38\n",
      "Current adversarial test loss: 1.8420692682266235\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.4\n",
      "Current benign test loss: 1.7361388206481934\n",
      "Current adversarial test accuracy: 0.37\n",
      "Current adversarial test loss: 1.9709980487823486\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.36\n",
      "Current benign test loss: 1.6480964422225952\n",
      "Current adversarial test accuracy: 0.26\n",
      "Current adversarial test loss: 1.8454499244689941\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.46\n",
      "Current benign test loss: 1.6233139038085938\n",
      "Current adversarial test accuracy: 0.35\n",
      "Current adversarial test loss: 1.848878026008606\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.29\n",
      "Current benign test loss: 1.7699487209320068\n",
      "Current adversarial test accuracy: 0.22\n",
      "Current adversarial test loss: 2.0400078296661377\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.36\n",
      "Current benign test loss: 1.7336790561676025\n",
      "Current adversarial test accuracy: 0.26\n",
      "Current adversarial test loss: 2.010291814804077\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.33\n",
      "Current benign test loss: 1.698716402053833\n",
      "Current adversarial test accuracy: 0.24\n",
      "Current adversarial test loss: 1.9063255786895752\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.31\n",
      "Current benign test loss: 1.738814353942871\n",
      "Current adversarial test accuracy: 0.24\n",
      "Current adversarial test loss: 2.006929874420166\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.45\n",
      "Current benign test loss: 1.6691484451293945\n",
      "Current adversarial test accuracy: 0.31\n",
      "Current adversarial test loss: 1.8947852849960327\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.35\n",
      "Current benign test loss: 1.6818585395812988\n",
      "Current adversarial test accuracy: 0.25\n",
      "Current adversarial test loss: 1.9049180746078491\n",
      "\n",
      "Total benign test accuarcy: 38.97\n",
      "Total adversarial test Accuarcy: 28.15\n",
      "Total benign test loss: 169.8468623161316\n",
      "Total adversarial test loss: 192.61357080936432\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 2 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.234375\n",
      "Current adversarial train loss: 1.9565144777297974\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.2578125\n",
      "Current adversarial train loss: 2.0047991275787354\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.28125\n",
      "Current adversarial train loss: 1.9711651802062988\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.21875\n",
      "Current adversarial train loss: 2.0768349170684814\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.2265625\n",
      "Current adversarial train loss: 2.004013776779175\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.25\n",
      "Current adversarial train loss: 1.9155869483947754\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.3046875\n",
      "Current adversarial train loss: 1.8973969221115112\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.3359375\n",
      "Current adversarial train loss: 1.8974964618682861\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.2421875\n",
      "Current adversarial train loss: 1.9781206846237183\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.3046875\n",
      "Current adversarial train loss: 1.9522473812103271\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.2578125\n",
      "Current adversarial train loss: 1.9638155698776245\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.25\n",
      "Current adversarial train loss: 1.8637285232543945\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.296875\n",
      "Current adversarial train loss: 1.8968228101730347\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.2265625\n",
      "Current adversarial train loss: 1.9819108247756958\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.28125\n",
      "Current adversarial train loss: 1.962427020072937\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.2578125\n",
      "Current adversarial train loss: 1.9062106609344482\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.2421875\n",
      "Current adversarial train loss: 1.967348575592041\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.234375\n",
      "Current adversarial train loss: 2.0053551197052\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.265625\n",
      "Current adversarial train loss: 2.0527143478393555\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.3359375\n",
      "Current adversarial train loss: 1.9150183200836182\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.3046875\n",
      "Current adversarial train loss: 1.9536232948303223\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.2421875\n",
      "Current adversarial train loss: 1.9990822076797485\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.3046875\n",
      "Current adversarial train loss: 1.8369064331054688\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.2734375\n",
      "Current adversarial train loss: 1.9158803224563599\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.21875\n",
      "Current adversarial train loss: 1.8791544437408447\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.25\n",
      "Current adversarial train loss: 1.9546165466308594\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.2890625\n",
      "Current adversarial train loss: 2.0218608379364014\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.28125\n",
      "Current adversarial train loss: 1.9219988584518433\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.21875\n",
      "Current adversarial train loss: 2.040151596069336\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.296875\n",
      "Current adversarial train loss: 1.8316770792007446\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.21875\n",
      "Current adversarial train loss: 2.0278754234313965\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.328125\n",
      "Current adversarial train loss: 1.8789409399032593\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.1953125\n",
      "Current adversarial train loss: 2.0152945518493652\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.3125\n",
      "Current adversarial train loss: 1.9098775386810303\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.25\n",
      "Current adversarial train loss: 2.0133743286132812\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.3984375\n",
      "Current adversarial train loss: 1.8214282989501953\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.265625\n",
      "Current adversarial train loss: 1.96546471118927\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.3125\n",
      "Current adversarial train loss: 1.9428696632385254\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.3359375\n",
      "Current adversarial train loss: 1.8436330556869507\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.2625\n",
      "Current adversarial train loss: 1.9106638431549072\n",
      "\n",
      "Total adversarial train accuarcy: 26.516\n",
      "Total adversarial train loss: 766.0000830888748\n",
      "\n",
      "[ Test epoch: 2 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.39\n",
      "Current benign test loss: 1.5825494527816772\n",
      "Current adversarial test accuracy: 0.33\n",
      "Current adversarial test loss: 1.840199589729309\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.46\n",
      "Current benign test loss: 1.6530505418777466\n",
      "Current adversarial test accuracy: 0.34\n",
      "Current adversarial test loss: 1.938571810722351\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.49\n",
      "Current benign test loss: 1.5461839437484741\n",
      "Current adversarial test accuracy: 0.33\n",
      "Current adversarial test loss: 1.7965409755706787\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.44\n",
      "Current benign test loss: 1.5265141725540161\n",
      "Current adversarial test accuracy: 0.34\n",
      "Current adversarial test loss: 1.7867038249969482\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.47\n",
      "Current benign test loss: 1.638823390007019\n",
      "Current adversarial test accuracy: 0.26\n",
      "Current adversarial test loss: 1.9454306364059448\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.42\n",
      "Current benign test loss: 1.6249990463256836\n",
      "Current adversarial test accuracy: 0.27\n",
      "Current adversarial test loss: 1.9310303926467896\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.41\n",
      "Current benign test loss: 1.5811489820480347\n",
      "Current adversarial test accuracy: 0.31\n",
      "Current adversarial test loss: 1.8408236503601074\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.43\n",
      "Current benign test loss: 1.6265885829925537\n",
      "Current adversarial test accuracy: 0.26\n",
      "Current adversarial test loss: 1.9648791551589966\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.45\n",
      "Current benign test loss: 1.570664644241333\n",
      "Current adversarial test accuracy: 0.34\n",
      "Current adversarial test loss: 1.8688631057739258\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.43\n",
      "Current benign test loss: 1.540718674659729\n",
      "Current adversarial test accuracy: 0.25\n",
      "Current adversarial test loss: 1.8033740520477295\n",
      "\n",
      "Total benign test accuarcy: 42.76\n",
      "Total adversarial test Accuarcy: 29.27\n",
      "Total benign test loss: 161.02358996868134\n",
      "Total adversarial test loss: 188.8143630027771\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 3 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.2890625\n",
      "Current adversarial train loss: 1.8459285497665405\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.25\n",
      "Current adversarial train loss: 1.9815726280212402\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.25\n",
      "Current adversarial train loss: 1.8851850032806396\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.3046875\n",
      "Current adversarial train loss: 1.9454069137573242\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.2578125\n",
      "Current adversarial train loss: 1.96026611328125\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.234375\n",
      "Current adversarial train loss: 1.9883321523666382\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.265625\n",
      "Current adversarial train loss: 1.9295177459716797\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.2734375\n",
      "Current adversarial train loss: 1.9248543977737427\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.265625\n",
      "Current adversarial train loss: 1.9204838275909424\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.2578125\n",
      "Current adversarial train loss: 1.9201695919036865\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.2578125\n",
      "Current adversarial train loss: 1.8008872270584106\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.2265625\n",
      "Current adversarial train loss: 1.9789187908172607\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.234375\n",
      "Current adversarial train loss: 1.8747559785842896\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.2890625\n",
      "Current adversarial train loss: 1.8632714748382568\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.2421875\n",
      "Current adversarial train loss: 1.969495415687561\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.3046875\n",
      "Current adversarial train loss: 1.955338716506958\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.3515625\n",
      "Current adversarial train loss: 1.8424615859985352\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.3515625\n",
      "Current adversarial train loss: 1.7937166690826416\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.328125\n",
      "Current adversarial train loss: 1.8707945346832275\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.25\n",
      "Current adversarial train loss: 1.9849013090133667\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.8450567722320557\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.296875\n",
      "Current adversarial train loss: 1.8919410705566406\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.34375\n",
      "Current adversarial train loss: 1.8875502347946167\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.25\n",
      "Current adversarial train loss: 1.996638536453247\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.2890625\n",
      "Current adversarial train loss: 1.9423465728759766\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.234375\n",
      "Current adversarial train loss: 2.028914451599121\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.2890625\n",
      "Current adversarial train loss: 1.8738459348678589\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.3515625\n",
      "Current adversarial train loss: 1.8547208309173584\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.28125\n",
      "Current adversarial train loss: 1.9136954545974731\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.3046875\n",
      "Current adversarial train loss: 1.863943338394165\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.2578125\n",
      "Current adversarial train loss: 1.9540565013885498\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.3359375\n",
      "Current adversarial train loss: 1.7850593328475952\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.3125\n",
      "Current adversarial train loss: 1.8637162446975708\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.25\n",
      "Current adversarial train loss: 2.0134074687957764\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.2578125\n",
      "Current adversarial train loss: 1.869649052619934\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.296875\n",
      "Current adversarial train loss: 1.948854684829712\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.25\n",
      "Current adversarial train loss: 1.9647624492645264\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.2421875\n",
      "Current adversarial train loss: 1.989176869392395\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.3046875\n",
      "Current adversarial train loss: 1.839961290359497\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.3375\n",
      "Current adversarial train loss: 1.797252893447876\n",
      "\n",
      "Total adversarial train accuarcy: 28.51\n",
      "Total adversarial train loss: 742.2918585538864\n",
      "\n",
      "[ Test epoch: 3 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.55\n",
      "Current benign test loss: 1.5110390186309814\n",
      "Current adversarial test accuracy: 0.36\n",
      "Current adversarial test loss: 1.8029682636260986\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.5\n",
      "Current benign test loss: 1.5278902053833008\n",
      "Current adversarial test accuracy: 0.32\n",
      "Current adversarial test loss: 1.8632234334945679\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.45\n",
      "Current benign test loss: 1.500819206237793\n",
      "Current adversarial test accuracy: 0.33\n",
      "Current adversarial test loss: 1.7803667783737183\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.52\n",
      "Current benign test loss: 1.4314390420913696\n",
      "Current adversarial test accuracy: 0.31\n",
      "Current adversarial test loss: 1.7212921380996704\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.43\n",
      "Current benign test loss: 1.6038033962249756\n",
      "Current adversarial test accuracy: 0.24\n",
      "Current adversarial test loss: 1.9668174982070923\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.43\n",
      "Current benign test loss: 1.5741366147994995\n",
      "Current adversarial test accuracy: 0.26\n",
      "Current adversarial test loss: 1.9302375316619873\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.36\n",
      "Current benign test loss: 1.5722169876098633\n",
      "Current adversarial test accuracy: 0.23\n",
      "Current adversarial test loss: 1.8755258321762085\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.45\n",
      "Current benign test loss: 1.5194202661514282\n",
      "Current adversarial test accuracy: 0.25\n",
      "Current adversarial test loss: 1.916764736175537\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.51\n",
      "Current benign test loss: 1.5231807231903076\n",
      "Current adversarial test accuracy: 0.36\n",
      "Current adversarial test loss: 1.8701655864715576\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.4\n",
      "Current benign test loss: 1.4852317571640015\n",
      "Current adversarial test accuracy: 0.21\n",
      "Current adversarial test loss: 1.7999200820922852\n",
      "\n",
      "Total benign test accuarcy: 46.26\n",
      "Total adversarial test Accuarcy: 30.35\n",
      "Total benign test loss: 152.37476527690887\n",
      "Total adversarial test loss: 183.84214079380035\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 4 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.265625\n",
      "Current adversarial train loss: 1.9234718084335327\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.2890625\n",
      "Current adversarial train loss: 1.771101474761963\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.3359375\n",
      "Current adversarial train loss: 1.8941211700439453\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.2734375\n",
      "Current adversarial train loss: 1.9438556432724\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.2890625\n",
      "Current adversarial train loss: 1.917062520980835\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.2578125\n",
      "Current adversarial train loss: 1.893394112586975\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.3125\n",
      "Current adversarial train loss: 1.8300790786743164\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.296875\n",
      "Current adversarial train loss: 1.8988966941833496\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.2734375\n",
      "Current adversarial train loss: 1.8314497470855713\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.3359375\n",
      "Current adversarial train loss: 1.8584233522415161\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.328125\n",
      "Current adversarial train loss: 1.7968919277191162\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.3359375\n",
      "Current adversarial train loss: 1.8276948928833008\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.3125\n",
      "Current adversarial train loss: 1.8725987672805786\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.2109375\n",
      "Current adversarial train loss: 1.9371041059494019\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.296875\n",
      "Current adversarial train loss: 1.948706030845642\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.328125\n",
      "Current adversarial train loss: 1.8693065643310547\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.2890625\n",
      "Current adversarial train loss: 1.8688665628433228\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.328125\n",
      "Current adversarial train loss: 1.7905738353729248\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.3203125\n",
      "Current adversarial train loss: 1.813489556312561\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.28125\n",
      "Current adversarial train loss: 1.8275694847106934\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.296875\n",
      "Current adversarial train loss: 1.8920838832855225\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.359375\n",
      "Current adversarial train loss: 1.8021948337554932\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.3125\n",
      "Current adversarial train loss: 1.8440532684326172\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.25\n",
      "Current adversarial train loss: 1.8465056419372559\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.2890625\n",
      "Current adversarial train loss: 1.8705974817276\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.3203125\n",
      "Current adversarial train loss: 1.76515793800354\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.328125\n",
      "Current adversarial train loss: 1.9002164602279663\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.3203125\n",
      "Current adversarial train loss: 1.7354975938796997\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.25\n",
      "Current adversarial train loss: 1.9742175340652466\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.328125\n",
      "Current adversarial train loss: 1.8602722883224487\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.328125\n",
      "Current adversarial train loss: 1.8701679706573486\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.3203125\n",
      "Current adversarial train loss: 1.8862154483795166\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.3125\n",
      "Current adversarial train loss: 1.8726732730865479\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.2734375\n",
      "Current adversarial train loss: 1.8896656036376953\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.3203125\n",
      "Current adversarial train loss: 1.8480472564697266\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.203125\n",
      "Current adversarial train loss: 1.9850077629089355\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.3203125\n",
      "Current adversarial train loss: 1.8241559267044067\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.25\n",
      "Current adversarial train loss: 1.9001314640045166\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.28125\n",
      "Current adversarial train loss: 1.8843176364898682\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.1875\n",
      "Current adversarial train loss: 1.9618871212005615\n",
      "\n",
      "Total adversarial train accuarcy: 30.262\n",
      "Total adversarial train loss: 722.773319363594\n",
      "\n",
      "[ Test epoch: 4 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.61\n",
      "Current benign test loss: 1.4061964750289917\n",
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.752732276916504\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.59\n",
      "Current benign test loss: 1.444547414779663\n",
      "Current adversarial test accuracy: 0.4\n",
      "Current adversarial test loss: 1.8041114807128906\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.42\n",
      "Current benign test loss: 1.404130458831787\n",
      "Current adversarial test accuracy: 0.3\n",
      "Current adversarial test loss: 1.695481777191162\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.55\n",
      "Current benign test loss: 1.3508769273757935\n",
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.6612708568572998\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.44\n",
      "Current benign test loss: 1.5110955238342285\n",
      "Current adversarial test accuracy: 0.28\n",
      "Current adversarial test loss: 1.9318950176239014\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.42\n",
      "Current benign test loss: 1.501359462738037\n",
      "Current adversarial test accuracy: 0.31\n",
      "Current adversarial test loss: 1.8958041667938232\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.39\n",
      "Current benign test loss: 1.447192668914795\n",
      "Current adversarial test accuracy: 0.29\n",
      "Current adversarial test loss: 1.745882511138916\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.49\n",
      "Current benign test loss: 1.4702545404434204\n",
      "Current adversarial test accuracy: 0.26\n",
      "Current adversarial test loss: 1.8853715658187866\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.5\n",
      "Current benign test loss: 1.4571980237960815\n",
      "Current adversarial test accuracy: 0.37\n",
      "Current adversarial test loss: 1.8123565912246704\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.44\n",
      "Current benign test loss: 1.4146875143051147\n",
      "Current adversarial test accuracy: 0.26\n",
      "Current adversarial test loss: 1.7559717893600464\n",
      "\n",
      "Total benign test accuarcy: 48.44\n",
      "Total adversarial test Accuarcy: 32.06\n",
      "Total benign test loss: 146.09417974948883\n",
      "Total adversarial test loss: 180.7355796098709\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 5 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.2890625\n",
      "Current adversarial train loss: 1.8692545890808105\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.7488645315170288\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.25\n",
      "Current adversarial train loss: 1.9005399942398071\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.3125\n",
      "Current adversarial train loss: 1.8446197509765625\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.3046875\n",
      "Current adversarial train loss: 1.9281600713729858\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.3203125\n",
      "Current adversarial train loss: 1.7186110019683838\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.34375\n",
      "Current adversarial train loss: 1.813575029373169\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.3515625\n",
      "Current adversarial train loss: 1.7447328567504883\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.6938462257385254\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.265625\n",
      "Current adversarial train loss: 1.900952696800232\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.3203125\n",
      "Current adversarial train loss: 1.727195143699646\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.3984375\n",
      "Current adversarial train loss: 1.728214144706726\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.296875\n",
      "Current adversarial train loss: 1.8257551193237305\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.2734375\n",
      "Current adversarial train loss: 1.86726975440979\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.1953125\n",
      "Current adversarial train loss: 1.8659952878952026\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.2890625\n",
      "Current adversarial train loss: 1.7756876945495605\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.3046875\n",
      "Current adversarial train loss: 1.8520889282226562\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.3984375\n",
      "Current adversarial train loss: 1.6567460298538208\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.3203125\n",
      "Current adversarial train loss: 1.829179048538208\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.3046875\n",
      "Current adversarial train loss: 1.779720425605774\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.3125\n",
      "Current adversarial train loss: 1.8522449731826782\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.296875\n",
      "Current adversarial train loss: 1.8650283813476562\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.28125\n",
      "Current adversarial train loss: 1.8856220245361328\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.6197892427444458\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.3203125\n",
      "Current adversarial train loss: 1.7221587896347046\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.3515625\n",
      "Current adversarial train loss: 1.7727266550064087\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.296875\n",
      "Current adversarial train loss: 1.8703213930130005\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.3515625\n",
      "Current adversarial train loss: 1.7018681764602661\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.3984375\n",
      "Current adversarial train loss: 1.7126877307891846\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.328125\n",
      "Current adversarial train loss: 1.7255809307098389\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.6624517440795898\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.3203125\n",
      "Current adversarial train loss: 1.7562536001205444\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.3125\n",
      "Current adversarial train loss: 1.851102352142334\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.3359375\n",
      "Current adversarial train loss: 1.8015272617340088\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.2578125\n",
      "Current adversarial train loss: 1.9365720748901367\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.3125\n",
      "Current adversarial train loss: 1.8404017686843872\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.3125\n",
      "Current adversarial train loss: 1.8195112943649292\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.28125\n",
      "Current adversarial train loss: 1.8117613792419434\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.7924541234970093\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.325\n",
      "Current adversarial train loss: 1.796332597732544\n",
      "\n",
      "Total adversarial train accuarcy: 32.094\n",
      "Total adversarial train loss: 704.5149304866791\n",
      "\n",
      "[ Test epoch: 5 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.58\n",
      "Current benign test loss: 1.3879507780075073\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.7402371168136597\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.53\n",
      "Current benign test loss: 1.39833664894104\n",
      "Current adversarial test accuracy: 0.4\n",
      "Current adversarial test loss: 1.7555580139160156\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.54\n",
      "Current benign test loss: 1.3222843408584595\n",
      "Current adversarial test accuracy: 0.38\n",
      "Current adversarial test loss: 1.6309398412704468\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.55\n",
      "Current benign test loss: 1.3154155015945435\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.6419248580932617\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.48\n",
      "Current benign test loss: 1.4082919359207153\n",
      "Current adversarial test accuracy: 0.3\n",
      "Current adversarial test loss: 1.8141984939575195\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.52\n",
      "Current benign test loss: 1.403141736984253\n",
      "Current adversarial test accuracy: 0.33\n",
      "Current adversarial test loss: 1.7917425632476807\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.47\n",
      "Current benign test loss: 1.3837422132492065\n",
      "Current adversarial test accuracy: 0.3\n",
      "Current adversarial test loss: 1.7042393684387207\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.49\n",
      "Current benign test loss: 1.4440008401870728\n",
      "Current adversarial test accuracy: 0.33\n",
      "Current adversarial test loss: 1.8501503467559814\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.51\n",
      "Current benign test loss: 1.386751413345337\n",
      "Current adversarial test accuracy: 0.35\n",
      "Current adversarial test loss: 1.763547420501709\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.46\n",
      "Current benign test loss: 1.3525207042694092\n",
      "Current adversarial test accuracy: 0.31\n",
      "Current adversarial test loss: 1.6967276334762573\n",
      "\n",
      "Total benign test accuarcy: 51.44\n",
      "Total adversarial test Accuarcy: 34.69\n",
      "Total benign test loss: 138.607076048851\n",
      "Total adversarial test loss: 173.96196103096008\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 6 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.643135666847229\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.8011832237243652\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.627785086631775\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.774290919303894\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.3359375\n",
      "Current adversarial train loss: 1.7629715204238892\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.7063158750534058\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.234375\n",
      "Current adversarial train loss: 1.9060771465301514\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.3359375\n",
      "Current adversarial train loss: 1.774608850479126\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.7654684782028198\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.7744128704071045\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.328125\n",
      "Current adversarial train loss: 1.7728257179260254\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.6691057682037354\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.7871942520141602\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.28125\n",
      "Current adversarial train loss: 1.8465982675552368\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.328125\n",
      "Current adversarial train loss: 1.7297263145446777\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.328125\n",
      "Current adversarial train loss: 1.7968933582305908\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.724799394607544\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.3515625\n",
      "Current adversarial train loss: 1.657551884651184\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.328125\n",
      "Current adversarial train loss: 1.7030402421951294\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.3125\n",
      "Current adversarial train loss: 1.745452642440796\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.3984375\n",
      "Current adversarial train loss: 1.6973819732666016\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.3515625\n",
      "Current adversarial train loss: 1.7993253469467163\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.265625\n",
      "Current adversarial train loss: 1.9565919637680054\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.359375\n",
      "Current adversarial train loss: 1.8163199424743652\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.3203125\n",
      "Current adversarial train loss: 1.659264326095581\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.3515625\n",
      "Current adversarial train loss: 1.7632088661193848\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.3203125\n",
      "Current adversarial train loss: 1.7714734077453613\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.7431803941726685\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.3515625\n",
      "Current adversarial train loss: 1.8167349100112915\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.3203125\n",
      "Current adversarial train loss: 1.7540814876556396\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.743574857711792\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.6183987855911255\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.7441272735595703\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.3984375\n",
      "Current adversarial train loss: 1.6496998071670532\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.7303112745285034\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.3046875\n",
      "Current adversarial train loss: 1.7846027612686157\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.3515625\n",
      "Current adversarial train loss: 1.782811164855957\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.328125\n",
      "Current adversarial train loss: 1.6547647714614868\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.296875\n",
      "Current adversarial train loss: 1.7606112957000732\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.35\n",
      "Current adversarial train loss: 1.8536770343780518\n",
      "\n",
      "Total adversarial train accuarcy: 33.954\n",
      "Total adversarial train loss: 684.6818844079971\n",
      "\n",
      "[ Test epoch: 6 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.58\n",
      "Current benign test loss: 1.2937601804733276\n",
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.7486680746078491\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.55\n",
      "Current benign test loss: 1.2927584648132324\n",
      "Current adversarial test accuracy: 0.4\n",
      "Current adversarial test loss: 1.7333238124847412\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.62\n",
      "Current benign test loss: 1.2000139951705933\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.5670928955078125\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.54\n",
      "Current benign test loss: 1.2166647911071777\n",
      "Current adversarial test accuracy: 0.32\n",
      "Current adversarial test loss: 1.6233989000320435\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.53\n",
      "Current benign test loss: 1.2371292114257812\n",
      "Current adversarial test accuracy: 0.37\n",
      "Current adversarial test loss: 1.7192730903625488\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.56\n",
      "Current benign test loss: 1.3033175468444824\n",
      "Current adversarial test accuracy: 0.34\n",
      "Current adversarial test loss: 1.7834210395812988\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.49\n",
      "Current benign test loss: 1.2872484922409058\n",
      "Current adversarial test accuracy: 0.3\n",
      "Current adversarial test loss: 1.6884026527404785\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.52\n",
      "Current benign test loss: 1.2608976364135742\n",
      "Current adversarial test accuracy: 0.34\n",
      "Current adversarial test loss: 1.7736315727233887\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.48\n",
      "Current benign test loss: 1.3298062086105347\n",
      "Current adversarial test accuracy: 0.31\n",
      "Current adversarial test loss: 1.7793809175491333\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.54\n",
      "Current benign test loss: 1.2628390789031982\n",
      "Current adversarial test accuracy: 0.33\n",
      "Current adversarial test loss: 1.6959948539733887\n",
      "\n",
      "Total benign test accuarcy: 53.49\n",
      "Total adversarial test Accuarcy: 34.07\n",
      "Total benign test loss: 128.81088209152222\n",
      "Total adversarial test loss: 172.7785609960556\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 7 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.6354409456253052\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.296875\n",
      "Current adversarial train loss: 1.829561471939087\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.7698389291763306\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.6232165098190308\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.3046875\n",
      "Current adversarial train loss: 1.7395179271697998\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.6558806896209717\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.6897447109222412\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.3359375\n",
      "Current adversarial train loss: 1.6962603330612183\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.6882036924362183\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.34375\n",
      "Current adversarial train loss: 1.8779473304748535\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.359375\n",
      "Current adversarial train loss: 1.6624754667282104\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.5886274576187134\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.3984375\n",
      "Current adversarial train loss: 1.7329198122024536\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.6470454931259155\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.7622507810592651\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.2421875\n",
      "Current adversarial train loss: 1.8242238759994507\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.265625\n",
      "Current adversarial train loss: 1.9186716079711914\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.328125\n",
      "Current adversarial train loss: 1.675661325454712\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.6660672426223755\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.3515625\n",
      "Current adversarial train loss: 1.718544602394104\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.6902354955673218\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.3515625\n",
      "Current adversarial train loss: 1.791252851486206\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.6776989698410034\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.6213140487670898\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.648886799812317\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.551539659500122\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.6486966609954834\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.6764836311340332\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.7440617084503174\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.3984375\n",
      "Current adversarial train loss: 1.6648521423339844\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.328125\n",
      "Current adversarial train loss: 1.6937228441238403\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.359375\n",
      "Current adversarial train loss: 1.7876778841018677\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.359375\n",
      "Current adversarial train loss: 1.6002612113952637\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.6698362827301025\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.2890625\n",
      "Current adversarial train loss: 1.9660159349441528\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.6371630430221558\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.6201255321502686\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.359375\n",
      "Current adversarial train loss: 1.7091598510742188\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.34375\n",
      "Current adversarial train loss: 1.6873424053192139\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.3375\n",
      "Current adversarial train loss: 1.6937599182128906\n",
      "\n",
      "Total adversarial train accuarcy: 35.578\n",
      "Total adversarial train loss: 667.1771539449692\n",
      "\n",
      "[ Test epoch: 7 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 1.255272388458252\n",
      "Current adversarial test accuracy: 0.36\n",
      "Current adversarial test loss: 1.655086874961853\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.6\n",
      "Current benign test loss: 1.2635393142700195\n",
      "Current adversarial test accuracy: 0.42\n",
      "Current adversarial test loss: 1.6902426481246948\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.59\n",
      "Current benign test loss: 1.1843862533569336\n",
      "Current adversarial test accuracy: 0.34\n",
      "Current adversarial test loss: 1.5384842157363892\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.61\n",
      "Current benign test loss: 1.1903982162475586\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.5719892978668213\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 1.2390568256378174\n",
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.6774466037750244\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.57\n",
      "Current benign test loss: 1.242741584777832\n",
      "Current adversarial test accuracy: 0.38\n",
      "Current adversarial test loss: 1.6640303134918213\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.58\n",
      "Current benign test loss: 1.2502995729446411\n",
      "Current adversarial test accuracy: 0.36\n",
      "Current adversarial test loss: 1.6222213506698608\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.62\n",
      "Current benign test loss: 1.180554986000061\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.6395831108093262\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.61\n",
      "Current benign test loss: 1.2503892183303833\n",
      "Current adversarial test accuracy: 0.36\n",
      "Current adversarial test loss: 1.6839653253555298\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.57\n",
      "Current benign test loss: 1.1761832237243652\n",
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.5863686800003052\n",
      "\n",
      "Total benign test accuarcy: 58.7\n",
      "Total adversarial test Accuarcy: 37.31\n",
      "Total benign test loss: 125.06811046600342\n",
      "Total adversarial test loss: 165.7786773443222\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 8 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.626987338066101\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.359375\n",
      "Current adversarial train loss: 1.6740485429763794\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.359375\n",
      "Current adversarial train loss: 1.6924437284469604\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.7021175622940063\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.5939067602157593\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.55807363986969\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.3984375\n",
      "Current adversarial train loss: 1.564272403717041\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.4910062551498413\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.3046875\n",
      "Current adversarial train loss: 1.7927688360214233\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.5835137367248535\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.34375\n",
      "Current adversarial train loss: 1.6444615125656128\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.359375\n",
      "Current adversarial train loss: 1.552634596824646\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.34375\n",
      "Current adversarial train loss: 1.6748466491699219\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.359375\n",
      "Current adversarial train loss: 1.6071103811264038\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.5430775880813599\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.3515625\n",
      "Current adversarial train loss: 1.6646206378936768\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.359375\n",
      "Current adversarial train loss: 1.707901120185852\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.6836295127868652\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.6782640218734741\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.6760116815567017\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.6351152658462524\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.6305034160614014\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.3125\n",
      "Current adversarial train loss: 1.6694567203521729\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.6245611906051636\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.6978294849395752\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.328125\n",
      "Current adversarial train loss: 1.7200111150741577\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.6076328754425049\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.359375\n",
      "Current adversarial train loss: 1.6185466051101685\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.5954999923706055\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.674645185470581\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.5528408288955688\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.3515625\n",
      "Current adversarial train loss: 1.7128276824951172\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.580643892288208\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.3984375\n",
      "Current adversarial train loss: 1.6233454942703247\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.5096726417541504\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.657986044883728\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.34375\n",
      "Current adversarial train loss: 1.6516607999801636\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.328125\n",
      "Current adversarial train loss: 1.6949312686920166\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.34375\n",
      "Current adversarial train loss: 1.685177206993103\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.4\n",
      "Current adversarial train loss: 1.5439153909683228\n",
      "\n",
      "Total adversarial train accuarcy: 36.984\n",
      "Total adversarial train loss: 650.2564452886581\n",
      "\n",
      "[ Test epoch: 8 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.61\n",
      "Current benign test loss: 1.1986712217330933\n",
      "Current adversarial test accuracy: 0.42\n",
      "Current adversarial test loss: 1.608442783355713\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.58\n",
      "Current benign test loss: 1.2258872985839844\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.6713414192199707\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.64\n",
      "Current benign test loss: 1.1710205078125\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.5399435758590698\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.58\n",
      "Current benign test loss: 1.141517996788025\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.5092799663543701\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.59\n",
      "Current benign test loss: 1.2045197486877441\n",
      "Current adversarial test accuracy: 0.34\n",
      "Current adversarial test loss: 1.6728456020355225\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.57\n",
      "Current benign test loss: 1.2403302192687988\n",
      "Current adversarial test accuracy: 0.35\n",
      "Current adversarial test loss: 1.690578579902649\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.56\n",
      "Current benign test loss: 1.2453479766845703\n",
      "Current adversarial test accuracy: 0.36\n",
      "Current adversarial test loss: 1.638819932937622\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.58\n",
      "Current benign test loss: 1.2322452068328857\n",
      "Current adversarial test accuracy: 0.37\n",
      "Current adversarial test loss: 1.7181118726730347\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.59\n",
      "Current benign test loss: 1.201263427734375\n",
      "Current adversarial test accuracy: 0.4\n",
      "Current adversarial test loss: 1.621955156326294\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.53\n",
      "Current benign test loss: 1.2071928977966309\n",
      "Current adversarial test accuracy: 0.36\n",
      "Current adversarial test loss: 1.6194381713867188\n",
      "\n",
      "Total benign test accuarcy: 58.42\n",
      "Total adversarial test Accuarcy: 39.11\n",
      "Total benign test loss: 120.59410464763641\n",
      "Total adversarial test loss: 162.32639920711517\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 9 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.557559847831726\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.8309848308563232\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.7263039350509644\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.6068198680877686\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.5154380798339844\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.6834062337875366\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.5622142553329468\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.6334201097488403\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.6824208498001099\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.5807507038116455\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.359375\n",
      "Current adversarial train loss: 1.6625747680664062\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.5558723211288452\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.34375\n",
      "Current adversarial train loss: 1.597953200340271\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.6015636920928955\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.3203125\n",
      "Current adversarial train loss: 1.7043911218643188\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.359375\n",
      "Current adversarial train loss: 1.6794791221618652\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.5157694816589355\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.6221413612365723\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.6125884056091309\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.6092582941055298\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.359375\n",
      "Current adversarial train loss: 1.7339167594909668\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.328125\n",
      "Current adversarial train loss: 1.6611063480377197\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.653283953666687\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.6243566274642944\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.5444029569625854\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.5940932035446167\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.6294232606887817\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.3515625\n",
      "Current adversarial train loss: 1.52193021774292\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.5612525939941406\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.574994683265686\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.3046875\n",
      "Current adversarial train loss: 1.6914730072021484\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.6020768880844116\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.662630319595337\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.611809253692627\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.2890625\n",
      "Current adversarial train loss: 1.773676872253418\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.646903395652771\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.5678915977478027\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.505402684211731\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.6206384897232056\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.4125\n",
      "Current adversarial train loss: 1.5736794471740723\n",
      "\n",
      "Total adversarial train accuarcy: 38.326\n",
      "Total adversarial train loss: 636.2274435758591\n",
      "\n",
      "[ Test epoch: 9 ]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.65\n",
      "Current benign test loss: 1.1091605424880981\n",
      "Current adversarial test accuracy: 0.37\n",
      "Current adversarial test loss: 1.6207597255706787\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.66\n",
      "Current benign test loss: 1.1269747018814087\n",
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.666967511177063\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.6\n",
      "Current benign test loss: 1.071142315864563\n",
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.523838996887207\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.66\n",
      "Current benign test loss: 1.0697647333145142\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.51837158203125\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.62\n",
      "Current benign test loss: 1.2164243459701538\n",
      "Current adversarial test accuracy: 0.3\n",
      "Current adversarial test loss: 1.8155752420425415\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.64\n",
      "Current benign test loss: 1.1626547574996948\n",
      "Current adversarial test accuracy: 0.33\n",
      "Current adversarial test loss: 1.7034629583358765\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 1.1238664388656616\n",
      "Current adversarial test accuracy: 0.35\n",
      "Current adversarial test loss: 1.6106703281402588\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.61\n",
      "Current benign test loss: 1.1422559022903442\n",
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.717667579650879\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.61\n",
      "Current benign test loss: 1.1253877878189087\n",
      "Current adversarial test accuracy: 0.33\n",
      "Current adversarial test loss: 1.6645095348358154\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.54\n",
      "Current benign test loss: 1.0820927619934082\n",
      "Current adversarial test accuracy: 0.34\n",
      "Current adversarial test loss: 1.5578113794326782\n",
      "\n",
      "Total benign test accuarcy: 61.73\n",
      "Total adversarial test Accuarcy: 37.33\n",
      "Total benign test loss: 113.04760026931763\n",
      "Total adversarial test loss: 164.59580862522125\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 10 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.6117771863937378\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.3359375\n",
      "Current adversarial train loss: 1.596998691558838\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.522424578666687\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.5510706901550293\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.34375\n",
      "Current adversarial train loss: 1.6259835958480835\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.5881527662277222\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.5600568056106567\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.5567985773086548\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.573641300201416\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.5016670227050781\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.5985620021820068\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.56475031375885\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.359375\n",
      "Current adversarial train loss: 1.694600224494934\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.6128820180892944\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.6574454307556152\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.6360831260681152\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.3515625\n",
      "Current adversarial train loss: 1.6902484893798828\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.5756330490112305\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.5972312688827515\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.6026355028152466\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.328125\n",
      "Current adversarial train loss: 1.601622223854065\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.3359375\n",
      "Current adversarial train loss: 1.6470474004745483\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.4601805210113525\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.5040239095687866\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.6562145948410034\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.558467984199524\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.5276589393615723\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.5652481317520142\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.5493441820144653\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.6079739332199097\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.464037537574768\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.3515625\n",
      "Current adversarial train loss: 1.6235733032226562\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.417343020439148\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.5561273097991943\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.3203125\n",
      "Current adversarial train loss: 1.695007085800171\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.6707069873809814\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.6679569482803345\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.6829060316085815\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.3359375\n",
      "Current adversarial train loss: 1.6277707815170288\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.4125\n",
      "Current adversarial train loss: 1.5738282203674316\n",
      "\n",
      "Total adversarial train accuarcy: 39.584\n",
      "Total adversarial train loss: 622.7054319381714\n",
      "\n",
      "[ Test epoch: 10 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 1.0900475978851318\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.628016710281372\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.65\n",
      "Current benign test loss: 1.1468859910964966\n",
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.7087101936340332\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.56\n",
      "Current benign test loss: 1.057325005531311\n",
      "Current adversarial test accuracy: 0.35\n",
      "Current adversarial test loss: 1.521360158920288\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.67\n",
      "Current benign test loss: 1.0794354677200317\n",
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.536775827407837\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.62\n",
      "Current benign test loss: 1.1228610277175903\n",
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.6816989183425903\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.61\n",
      "Current benign test loss: 1.1135334968566895\n",
      "Current adversarial test accuracy: 0.4\n",
      "Current adversarial test loss: 1.652863621711731\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.63\n",
      "Current benign test loss: 1.1423661708831787\n",
      "Current adversarial test accuracy: 0.33\n",
      "Current adversarial test loss: 1.6322575807571411\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.64\n",
      "Current benign test loss: 1.1002628803253174\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Current adversarial test accuracy: 0.38\n",
      "Current adversarial test loss: 1.6856613159179688\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.64\n",
      "Current benign test loss: 1.0707169771194458\n",
      "Current adversarial test accuracy: 0.34\n",
      "Current adversarial test loss: 1.6210136413574219\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.69\n",
      "Current benign test loss: 1.0604130029678345\n",
      "Current adversarial test accuracy: 0.42\n",
      "Current adversarial test loss: 1.5566291809082031\n",
      "\n",
      "Total benign test accuarcy: 64.05\n",
      "Total adversarial test Accuarcy: 39.48\n",
      "Total benign test loss: 110.10043334960938\n",
      "Total adversarial test loss: 161.66028439998627\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 11 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.5967539548873901\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.559847354888916\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.3984375\n",
      "Current adversarial train loss: 1.6423707008361816\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.328125\n",
      "Current adversarial train loss: 1.6822510957717896\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.3203125\n",
      "Current adversarial train loss: 1.7375704050064087\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.5402816534042358\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.670456886291504\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.609628438949585\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.545539379119873\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.6543642282485962\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.5183435678482056\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.4588921070098877\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.623768925666809\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.5312721729278564\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.6094062328338623\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.4224908351898193\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.3515625\n",
      "Current adversarial train loss: 1.6417396068572998\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.5443110466003418\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.5642353296279907\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.515860915184021\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4518359899520874\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4480724334716797\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.486464500427246\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.5698659420013428\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.3984375\n",
      "Current adversarial train loss: 1.6727352142333984\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.5940806865692139\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.540199637413025\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.5521454811096191\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.6125564575195312\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.5195114612579346\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.402280569076538\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.3984375\n",
      "Current adversarial train loss: 1.5599905252456665\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.457398772239685\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.6447068452835083\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.7088994979858398\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.4661723375320435\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4316056966781616\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.6204931735992432\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.3984375\n",
      "Current adversarial train loss: 1.5820655822753906\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.4274367094039917\n",
      "\n",
      "Total adversarial train accuarcy: 40.938\n",
      "Total adversarial train loss: 611.0040677785873\n",
      "\n",
      "[ Test epoch: 11 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 1.0475560426712036\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.5765074491500854\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.63\n",
      "Current benign test loss: 1.0459160804748535\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.5939338207244873\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.65\n",
      "Current benign test loss: 1.0434054136276245\n",
      "Current adversarial test accuracy: 0.38\n",
      "Current adversarial test loss: 1.490622878074646\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.63\n",
      "Current benign test loss: 1.03708016872406\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.498367428779602\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.64\n",
      "Current benign test loss: 1.059311866760254\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.5741124153137207\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.62\n",
      "Current benign test loss: 1.058823823928833\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.5700747966766357\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.63\n",
      "Current benign test loss: 1.1068837642669678\n",
      "Current adversarial test accuracy: 0.37\n",
      "Current adversarial test loss: 1.5851240158081055\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.67\n",
      "Current benign test loss: 0.9710388779640198\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.5120837688446045\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.62\n",
      "Current benign test loss: 1.0589450597763062\n",
      "Current adversarial test accuracy: 0.4\n",
      "Current adversarial test loss: 1.6230387687683105\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.69\n",
      "Current benign test loss: 0.9806514978408813\n",
      "Current adversarial test accuracy: 0.4\n",
      "Current adversarial test loss: 1.4720449447631836\n",
      "\n",
      "Total benign test accuarcy: 65.03\n",
      "Total adversarial test Accuarcy: 41.17\n",
      "Total benign test loss: 104.37412399053574\n",
      "Total adversarial test loss: 155.48519909381866\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 12 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.5774036645889282\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4533803462982178\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.359375\n",
      "Current adversarial train loss: 1.6515241861343384\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.3359375\n",
      "Current adversarial train loss: 1.779713749885559\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.5693316459655762\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.5387194156646729\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.5619356632232666\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.5852153301239014\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.4904998540878296\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.4405243396759033\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.3984375\n",
      "Current adversarial train loss: 1.5668830871582031\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.5419130325317383\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.6323459148406982\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.3984375\n",
      "Current adversarial train loss: 1.5467675924301147\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.5685652494430542\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.443057894706726\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.3883146047592163\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.3515625\n",
      "Current adversarial train loss: 1.5563572645187378\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.4923598766326904\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.359375\n",
      "Current adversarial train loss: 1.6049025058746338\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.608148217201233\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.3203125\n",
      "Current adversarial train loss: 1.614335060119629\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.6083050966262817\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.5272055864334106\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.5261706113815308\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.5598424673080444\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.446102261543274\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.6371402740478516\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.3125\n",
      "Current adversarial train loss: 1.6355340480804443\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.3970658779144287\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.4432886838912964\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4725710153579712\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.429174542427063\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.5926042795181274\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.3984375\n",
      "Current adversarial train loss: 1.5701850652694702\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.485749363899231\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.46165132522583\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.5143964290618896\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.6070493459701538\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.3625\n",
      "Current adversarial train loss: 1.5859843492507935\n",
      "\n",
      "Total adversarial train accuarcy: 41.556\n",
      "Total adversarial train loss: 600.4303967952728\n",
      "\n",
      "[ Test epoch: 12 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.61\n",
      "Current benign test loss: 1.0792213678359985\n",
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.605710506439209\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.65\n",
      "Current benign test loss: 1.1121928691864014\n",
      "Current adversarial test accuracy: 0.38\n",
      "Current adversarial test loss: 1.6763124465942383\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.65\n",
      "Current benign test loss: 1.0337575674057007\n",
      "Current adversarial test accuracy: 0.4\n",
      "Current adversarial test loss: 1.4758203029632568\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.65\n",
      "Current benign test loss: 1.0895156860351562\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.5558476448059082\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 1.1163636445999146\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.6528064012527466\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.67\n",
      "Current benign test loss: 1.0741989612579346\n",
      "Current adversarial test accuracy: 0.4\n",
      "Current adversarial test loss: 1.6034557819366455\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.56\n",
      "Current benign test loss: 1.1357793807983398\n",
      "Current adversarial test accuracy: 0.31\n",
      "Current adversarial test loss: 1.636021375656128\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.63\n",
      "Current benign test loss: 1.0469639301300049\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.6258587837219238\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.63\n",
      "Current benign test loss: 1.0993715524673462\n",
      "Current adversarial test accuracy: 0.37\n",
      "Current adversarial test loss: 1.6351836919784546\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.58\n",
      "Current benign test loss: 1.082804560661316\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.5646549463272095\n",
      "\n",
      "Total benign test accuarcy: 63.78\n",
      "Total adversarial test Accuarcy: 39.79\n",
      "Total benign test loss: 108.61819666624069\n",
      "Total adversarial test loss: 159.33313035964966\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 13 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4193811416625977\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.4750381708145142\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.4992529153823853\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.4659719467163086\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.3984375\n",
      "Current adversarial train loss: 1.5296990871429443\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4781861305236816\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.587296724319458\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.5117546319961548\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.4681134223937988\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4493767023086548\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.4148168563842773\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.5366166830062866\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.428432822227478\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.4045360088348389\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.5809624195098877\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.5161665678024292\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.4650689363479614\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.359375\n",
      "Current adversarial train loss: 1.517251968383789\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.4842236042022705\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.5878894329071045\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.5099378824234009\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.4755902290344238\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.6258323192596436\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.479872226715088\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.510003685951233\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.6103817224502563\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.4171160459518433\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.34375\n",
      "Current adversarial train loss: 1.5856178998947144\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.5680292844772339\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.3984375\n",
      "Current adversarial train loss: 1.5262317657470703\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.4939996004104614\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.4621514081954956\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.5650666952133179\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.4954363107681274\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4283620119094849\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3631360530853271\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.359375\n",
      "Current adversarial train loss: 1.7167866230010986\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.5177477598190308\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.5298731327056885\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.4\n",
      "Current adversarial train loss: 1.5464873313903809\n",
      "\n",
      "Total adversarial train accuarcy: 42.582\n",
      "Total adversarial train loss: 591.2180813550949\n",
      "\n",
      "[ Test epoch: 13 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.71\n",
      "Current benign test loss: 1.0024679899215698\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.4978303909301758\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.990589439868927\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.490502119064331\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.6\n",
      "Current benign test loss: 0.9667354822158813\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.3882758617401123\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.71\n",
      "Current benign test loss: 0.9789637923240662\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.4270684719085693\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 0.993501603603363\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.5022399425506592\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 1.032887578010559\n",
      "Current adversarial test accuracy: 0.38\n",
      "Current adversarial test loss: 1.582685947418213\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.64\n",
      "Current benign test loss: 1.0577778816223145\n",
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.5306472778320312\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.66\n",
      "Current benign test loss: 0.9661896228790283\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.5082192420959473\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.66\n",
      "Current benign test loss: 1.0335760116577148\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.5556052923202515\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 0.9489964246749878\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.420934796333313\n",
      "\n",
      "Total benign test accuarcy: 68.63\n",
      "Total adversarial test Accuarcy: 43.14\n",
      "Total benign test loss: 100.05362010002136\n",
      "Total adversarial test loss: 149.7149224281311\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 14 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.5154756307601929\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.3848713636398315\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.568171501159668\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3681992292404175\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4146952629089355\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.3795729875564575\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.359375\n",
      "Current adversarial train loss: 1.5584118366241455\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.5029258728027344\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.5100226402282715\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.4173567295074463\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.4389675855636597\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4609308242797852\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.5044747591018677\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.5737303495407104\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.6158671379089355\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.3493516445159912\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.6475647687911987\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.445770263671875\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.6126105785369873\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.4024994373321533\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.488791584968567\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.5225027799606323\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2769973278045654\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.4806312322616577\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.3125\n",
      "Current adversarial train loss: 1.5527998208999634\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.5165331363677979\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4929227828979492\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.549132227897644\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.5989463329315186\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4817198514938354\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.5438023805618286\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.5888946056365967\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.4078692197799683\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.531705617904663\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3635871410369873\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.545470952987671\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.4779423475265503\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.3752378225326538\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.4988864660263062\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.4125\n",
      "Current adversarial train loss: 1.447906255722046\n",
      "\n",
      "Total adversarial train accuarcy: 43.07\n",
      "Total adversarial train loss: 584.1943645477295\n",
      "\n",
      "[ Test epoch: 14 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.75\n",
      "Current benign test loss: 0.9671921730041504\n",
      "Current adversarial test accuracy: 0.42\n",
      "Current adversarial test loss: 1.4266139268875122\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.67\n",
      "Current benign test loss: 1.0299005508422852\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.5512205362319946\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 0.9758642315864563\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.4034253358840942\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 0.9990198016166687\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.4604207277297974\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 0.9942814707756042\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.4848980903625488\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.66\n",
      "Current benign test loss: 1.008100986480713\n",
      "Current adversarial test accuracy: 0.4\n",
      "Current adversarial test loss: 1.5295714139938354\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.66\n",
      "Current benign test loss: 1.0743387937545776\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.5535380840301514\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 0.9382158517837524\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.4356387853622437\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.66\n",
      "Current benign test loss: 1.0160833597183228\n",
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.5288026332855225\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.64\n",
      "Current benign test loss: 1.0009238719940186\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.4916260242462158\n",
      "\n",
      "Total benign test accuarcy: 66.95\n",
      "Total adversarial test Accuarcy: 43.96\n",
      "Total benign test loss: 100.6215655207634\n",
      "Total adversarial test loss: 149.632364153862\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 15 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3415333032608032\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.3936853408813477\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.4859355688095093\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.5465961694717407\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.5311027765274048\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.382603406906128\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.4582087993621826\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.4204034805297852\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.3299126625061035\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3996634483337402\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.500333309173584\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.4997162818908691\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3419348001480103\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.5167747735977173\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.4436707496643066\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.5007439851760864\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.593754529953003\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.4463937282562256\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4555529356002808\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.243133544921875\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.529751181602478\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3561017513275146\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.4809125661849976\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.3986884355545044\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3539350032806396\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.3984375\n",
      "Current adversarial train loss: 1.5416715145111084\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.550763726234436\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.58692467212677\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.5244097709655762\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.4337960481643677\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.3564382791519165\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.5569758415222168\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.5086236000061035\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.4940215349197388\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.5191543102264404\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.49125075340271\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3039835691452026\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2729204893112183\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.4377219676971436\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.3625\n",
      "Current adversarial train loss: 1.6391088962554932\n",
      "\n",
      "Total adversarial train accuarcy: 44.336\n",
      "Total adversarial train loss: 574.223864197731\n",
      "\n",
      "[ Test epoch: 15 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.9904949069023132\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.5230330228805542\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 0.9862459301948547\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.5352717638015747\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.63\n",
      "Current benign test loss: 0.9882600903511047\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.440920352935791\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 0.9843924641609192\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.4706954956054688\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.69\n",
      "Current benign test loss: 1.0212152004241943\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.5738638639450073\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.69\n",
      "Current benign test loss: 1.0269285440444946\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.6125712394714355\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.65\n",
      "Current benign test loss: 1.0354443788528442\n",
      "Current adversarial test accuracy: 0.36\n",
      "Current adversarial test loss: 1.5260674953460693\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.9008187055587769\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.4343754053115845\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.9522838592529297\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.4988560676574707\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 0.9616140723228455\n",
      "Current adversarial test accuracy: 0.4\n",
      "Current adversarial test loss: 1.4684242010116577\n",
      "\n",
      "Total benign test accuarcy: 69.01\n",
      "Total adversarial test Accuarcy: 43.13\n",
      "Total benign test loss: 98.64405047893524\n",
      "Total adversarial test loss: 151.3996628522873\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 16 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.4367252588272095\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.507033348083496\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.4001612663269043\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.3444160223007202\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.456596851348877\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.4944207668304443\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.3772692680358887\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.336578130722046\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.3877604007720947\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3323496580123901\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.4016435146331787\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3694217205047607\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.5008420944213867\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3701558113098145\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.396401286125183\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.530869960784912\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.5102285146713257\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.4257559776306152\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.454073190689087\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.2623674869537354\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.4272034168243408\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.3092255592346191\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.2929682731628418\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.4338308572769165\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.3984375\n",
      "Current adversarial train loss: 1.4118633270263672\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.3702974319458008\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.549904704093933\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.4400756359100342\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.3682630062103271\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.6015625\n",
      "Current adversarial train loss: 1.272025465965271\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.493049144744873\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.3524341583251953\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3442150354385376\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.359375\n",
      "Current adversarial train loss: 1.5313878059387207\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.2621077299118042\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.6030691862106323\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.4001044034957886\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.3713045120239258\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.4153608083724976\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.575\n",
      "Current adversarial train loss: 1.2350022792816162\n",
      "\n",
      "Total adversarial train accuarcy: 45.03\n",
      "Total adversarial train loss: 566.0744944810867\n",
      "\n",
      "[ Test epoch: 16 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.9509263038635254\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.5121891498565674\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.71\n",
      "Current benign test loss: 0.9369211792945862\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.5060168504714966\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.64\n",
      "Current benign test loss: 0.9275122880935669\n",
      "Current adversarial test accuracy: 0.4\n",
      "Current adversarial test loss: 1.377396583557129\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.9256134033203125\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.390714406967163\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 0.9397956132888794\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.4691141843795776\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.79\n",
      "Current benign test loss: 0.9362136721611023\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.4957836866378784\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.65\n",
      "Current benign test loss: 1.0029298067092896\n",
      "Current adversarial test accuracy: 0.37\n",
      "Current adversarial test loss: 1.5210597515106201\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 0.8497560620307922\n",
      "Current adversarial test accuracy: 0.5\n",
      "Current adversarial test loss: 1.388840913772583\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.9408103227615356\n",
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.5345855951309204\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.71\n",
      "Current benign test loss: 0.9125533103942871\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.4260450601577759\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Total benign test accuarcy: 70.09\n",
      "Total adversarial test Accuarcy: 44.46\n",
      "Total benign test loss: 93.66005980968475\n",
      "Total adversarial test loss: 146.55031299591064\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 17 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.4048186540603638\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.305162787437439\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.420322060585022\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.3346128463745117\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4685219526290894\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.5792245864868164\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.439427375793457\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.4181702136993408\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.4333653450012207\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3979135751724243\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.348949670791626\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3608871698379517\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4615460634231567\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3346266746520996\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.294472336769104\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4431058168411255\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.5019662380218506\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.4049993753433228\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.4125367403030396\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.436320185661316\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.3024297952651978\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4245778322219849\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.6190588474273682\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3322088718414307\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.364859700202942\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.3825302124023438\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.4059251546859741\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.3187731504440308\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4240373373031616\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.418565034866333\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2952007055282593\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3677033185958862\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.4863530397415161\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.3954830169677734\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.4574552774429321\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.3984375\n",
      "Current adversarial train loss: 1.5025466680526733\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.448421835899353\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4947293996810913\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4683862924575806\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.4\n",
      "Current adversarial train loss: 1.4501068592071533\n",
      "\n",
      "Total adversarial train accuarcy: 45.36\n",
      "Total adversarial train loss: 560.5048512220383\n",
      "\n",
      "[ Test epoch: 17 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.75\n",
      "Current benign test loss: 0.8946412801742554\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.4124150276184082\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.71\n",
      "Current benign test loss: 0.9087474346160889\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.4895679950714111\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 0.9505489468574524\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.4041041135787964\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 0.931221604347229\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.402773380279541\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.9526974558830261\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.5406646728515625\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.78\n",
      "Current benign test loss: 0.9203891158103943\n",
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.4806629419326782\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 1.0093088150024414\n",
      "Current adversarial test accuracy: 0.35\n",
      "Current adversarial test loss: 1.52333664894104\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 0.8621937036514282\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.3988157510757446\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.71\n",
      "Current benign test loss: 0.8940658569335938\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.4914257526397705\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.886720597743988\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.4025496244430542\n",
      "\n",
      "Total benign test accuarcy: 70.43\n",
      "Total adversarial test Accuarcy: 44.79\n",
      "Total benign test loss: 92.29528468847275\n",
      "Total adversarial test loss: 146.25586223602295\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 18 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4815090894699097\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.474358081817627\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.4094429016113281\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2962833642959595\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.5042420625686646\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.3415898084640503\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.373828649520874\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.3797391653060913\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.429718255996704\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.3857108354568481\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3502122163772583\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.4756730794906616\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.4917103052139282\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.3565161228179932\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.3445793390274048\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3701486587524414\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.429449439048767\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.279382586479187\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4147350788116455\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.4917778968811035\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.4547137022018433\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.6123616695404053\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.2557731866836548\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2911375761032104\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.4689565896987915\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.264453649520874\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3152602910995483\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.4264291524887085\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.3718370199203491\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.3735021352767944\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.4899461269378662\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.359375\n",
      "Current adversarial train loss: 1.6168290376663208\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.3841975927352905\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.41441011428833\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.5108280181884766\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.5056586265563965\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4825955629348755\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.3883274793624878\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.334028959274292\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.378629207611084\n",
      "\n",
      "Total adversarial train accuarcy: 45.744\n",
      "Total adversarial train loss: 555.5075969696045\n",
      "\n",
      "[ Test epoch: 18 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.71\n",
      "Current benign test loss: 0.9414828419685364\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.458461046218872\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.9254181385040283\n",
      "Current adversarial test accuracy: 0.5\n",
      "Current adversarial test loss: 1.4763405323028564\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 0.9120123982429504\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.3837414979934692\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 0.9706568121910095\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.4906305074691772\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 1.0064376592636108\n",
      "Current adversarial test accuracy: 0.42\n",
      "Current adversarial test loss: 1.564608097076416\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.75\n",
      "Current benign test loss: 0.9782286286354065\n",
      "Current adversarial test accuracy: 0.4\n",
      "Current adversarial test loss: 1.5572068691253662\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.65\n",
      "Current benign test loss: 1.027698040008545\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.5257632732391357\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.57\n",
      "Current benign test loss: 0.9626491069793701\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.5632725954055786\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.9138734936714172\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.4604519605636597\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.67\n",
      "Current benign test loss: 0.9029150605201721\n",
      "Current adversarial test accuracy: 0.4\n",
      "Current adversarial test loss: 1.459367036819458\n",
      "\n",
      "Total benign test accuarcy: 69.03\n",
      "Total adversarial test Accuarcy: 43.96\n",
      "Total benign test loss: 94.02762019634247\n",
      "Total adversarial test loss: 147.831791639328\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 19 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.2811857461929321\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.3473199605941772\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.5169883966445923\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.2909756898880005\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3685433864593506\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.4628652334213257\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.425345540046692\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.3920869827270508\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.5307806730270386\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.5859375\n",
      "Current adversarial train loss: 1.283434271812439\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3115969896316528\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.2949570417404175\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4230164289474487\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.3892008066177368\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.6295186281204224\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.4639880657196045\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.396653175354004\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.4270858764648438\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3159770965576172\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3359252214431763\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.3953887224197388\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.5150346755981445\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4622151851654053\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3518574237823486\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3215456008911133\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.4715718030929565\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.4451860189437866\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4297044277191162\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2936416864395142\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.3942482471466064\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.4982874393463135\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.396478533744812\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.534975290298462\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3328111171722412\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.3218388557434082\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.4924366474151611\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.5058633089065552\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4617811441421509\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.3984261751174927\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4550530910491943\n",
      "\n",
      "Total adversarial train accuarcy: 46.652\n",
      "Total adversarial train loss: 548.8169894218445\n",
      "\n",
      "[ Test epoch: 19 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.912537157535553\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.456335186958313\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.8806888461112976\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.4496121406555176\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 0.9096513986587524\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.416340947151184\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.67\n",
      "Current benign test loss: 0.9322927594184875\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.4668333530426025\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.949934720993042\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.5426439046859741\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.9146717190742493\n",
      "Current adversarial test accuracy: 0.38\n",
      "Current adversarial test loss: 1.5042763948440552\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.64\n",
      "Current benign test loss: 0.9812664985656738\n",
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.5535343885421753\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.67\n",
      "Current benign test loss: 0.8826032876968384\n",
      "Current adversarial test accuracy: 0.5\n",
      "Current adversarial test loss: 1.4882103204727173\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.8752628564834595\n",
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.4553543329238892\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.8371261358261108\n",
      "Current adversarial test accuracy: 0.42\n",
      "Current adversarial test loss: 1.368466854095459\n",
      "\n",
      "Total benign test accuarcy: 70.34\n",
      "Total adversarial test Accuarcy: 44.56\n",
      "Total benign test loss: 90.78137022256851\n",
      "Total adversarial test loss: 146.75732100009918\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 20 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.521525263786316\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3523629903793335\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.4339025020599365\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4481927156448364\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.41715669631958\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.3576350212097168\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.5914196968078613\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.2838726043701172\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.3491733074188232\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.3607925176620483\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.548640251159668\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.4535633325576782\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.3539618253707886\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.3952149152755737\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.3671875\n",
      "Current adversarial train loss: 1.5894300937652588\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.4058375358581543\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.4232151508331299\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3448482751846313\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3562453985214233\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.1664117574691772\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.4374091625213623\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.4132317304611206\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.319601058959961\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.334679365158081\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.4298475980758667\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.3914051055908203\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.3102707862854004\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2303354740142822\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4062182903289795\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.3954256772994995\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.507383108139038\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4599522352218628\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.2081348896026611\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4892674684524536\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.4273015260696411\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.24685537815094\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.45957612991333\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.373092532157898\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.418512225151062\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.4875\n",
      "Current adversarial train loss: 1.5378119945526123\n",
      "\n",
      "Total adversarial train accuarcy: 46.98\n",
      "Total adversarial train loss: 544.0316988229752\n",
      "\n",
      "[ Test epoch: 20 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.877989649772644\n",
      "Current adversarial test accuracy: 0.51\n",
      "Current adversarial test loss: 1.4655779600143433\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.8583118915557861\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.4703059196472168\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.66\n",
      "Current benign test loss: 0.8670493960380554\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.3684303760528564\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.71\n",
      "Current benign test loss: 0.9270622134208679\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.4750369787216187\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 0.901985228061676\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.4640332460403442\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.8830441236495972\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.4756762981414795\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.65\n",
      "Current benign test loss: 0.9321513175964355\n",
      "Current adversarial test accuracy: 0.42\n",
      "Current adversarial test loss: 1.4845584630966187\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 0.815040111541748\n",
      "Current adversarial test accuracy: 0.52\n",
      "Current adversarial test loss: 1.3951122760772705\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.8499938249588013\n",
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.4588534832000732\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.64\n",
      "Current benign test loss: 0.8799548149108887\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.4452227354049683\n",
      "\n",
      "Total benign test accuarcy: 70.88\n",
      "Total adversarial test Accuarcy: 46.2\n",
      "Total benign test loss: 87.50093299150467\n",
      "Total adversarial test loss: 143.99810528755188\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 21 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.512281894683838\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.5717450380325317\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.3303860425949097\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.3665510416030884\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.3549860715866089\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3513644933700562\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.2773375511169434\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.2874109745025635\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.2448360919952393\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.3619226217269897\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.383352279663086\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.5036461353302002\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.3917497396469116\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.3605622053146362\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.3237857818603516\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.444130778312683\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.3450640439987183\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.396891474723816\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.3146096467971802\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.3276855945587158\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4373351335525513\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.359478235244751\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.345521330833435\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.4075775146484375\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.4210668802261353\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2430365085601807\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3389548063278198\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.413983941078186\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.3515048027038574\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.363661527633667\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.2995305061340332\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.5269547700881958\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.4333996772766113\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.4032734632492065\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.3988370895385742\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3052667379379272\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2500773668289185\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4115607738494873\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.2606498003005981\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.4875\n",
      "Current adversarial train loss: 1.3441630601882935\n",
      "\n",
      "Total adversarial train accuarcy: 47.196\n",
      "Total adversarial train loss: 539.8491376638412\n",
      "\n",
      "[ Test epoch: 21 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.78\n",
      "Current benign test loss: 0.8838927745819092\n",
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.4341472387313843\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.8538379073143005\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.438727855682373\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.9284583330154419\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.425957202911377\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 0.890689492225647\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.3983395099639893\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.9420568943023682\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.4779002666473389\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.9128112196922302\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.4868708848953247\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.67\n",
      "Current benign test loss: 0.9067482948303223\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.4270460605621338\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.71\n",
      "Current benign test loss: 0.821540117263794\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.3828113079071045\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.8487136363983154\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.4422972202301025\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.69\n",
      "Current benign test loss: 0.8606346249580383\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.4250565767288208\n",
      "\n",
      "Total benign test accuarcy: 72.44\n",
      "Total adversarial test Accuarcy: 45.43\n",
      "Total benign test loss: 88.90441638231277\n",
      "Total adversarial test loss: 144.07864153385162\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 22 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.4411075115203857\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.348787784576416\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.2714483737945557\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3151135444641113\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.3382892608642578\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.28641939163208\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.2889480590820312\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.469252109527588\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4558954238891602\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.4337283372879028\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.3230537176132202\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4032504558563232\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.3945894241333008\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.5610853433609009\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.3617172241210938\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.3516361713409424\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3015409708023071\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.340117335319519\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.3828474283218384\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.430063009262085\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.3940662145614624\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.2964173555374146\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.3568527698516846\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.301232933998108\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.3694599866867065\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.4568418264389038\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.327763557434082\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.5009379386901855\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3223412036895752\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.326454997062683\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2158743143081665\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.4456080198287964\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3095529079437256\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.4216428995132446\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.3852946758270264\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.388604760169983\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.6165391206741333\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.4061814546585083\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.29060697555542\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.3645457029342651\n",
      "\n",
      "Total adversarial train accuarcy: 47.46\n",
      "Total adversarial train loss: 535.691565990448\n",
      "\n",
      "[ Test epoch: 22 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.8672652244567871\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.4214434623718262\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.8655864596366882\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.4333527088165283\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.78\n",
      "Current benign test loss: 0.869171142578125\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.3677412271499634\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.8733077049255371\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.36126708984375\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.9028794765472412\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.432359218597412\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.85\n",
      "Current benign test loss: 0.9088729619979858\n",
      "Current adversarial test accuracy: 0.38\n",
      "Current adversarial test loss: 1.4698512554168701\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 0.9583437442779541\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.4682408571243286\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.69\n",
      "Current benign test loss: 0.8287675380706787\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.3880494832992554\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.76\n",
      "Current benign test loss: 0.8879646062850952\n",
      "Current adversarial test accuracy: 0.5\n",
      "Current adversarial test loss: 1.5037949085235596\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.76\n",
      "Current benign test loss: 0.8365169763565063\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.3622419834136963\n",
      "\n",
      "Total benign test accuarcy: 73.93\n",
      "Total adversarial test Accuarcy: 46.58\n",
      "Total benign test loss: 88.16376179456711\n",
      "Total adversarial test loss: 142.47517108917236\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 23 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.308409333229065\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.4243165254592896\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.4245346784591675\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.2487761974334717\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.1712870597839355\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3820890188217163\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.3497647047042847\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.2997701168060303\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.4409520626068115\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.2960271835327148\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.5669357776641846\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2988044023513794\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.230023980140686\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.3866362571716309\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4678784608840942\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.2696934938430786\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.3206835985183716\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.3714569807052612\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.400529146194458\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.5181406736373901\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.2519968748092651\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.3867814540863037\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3503316640853882\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3419415950775146\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.3695801496505737\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4212090969085693\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.3997256755828857\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.317546010017395\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.3992550373077393\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.3410218954086304\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.3829783201217651\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.400335669517517\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.407299280166626\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.2672609090805054\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.3532555103302002\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4025731086730957\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.2212251424789429\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.3984375\n",
      "Current adversarial train loss: 1.502030611038208\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.377970814704895\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.5375\n",
      "Current adversarial train loss: 1.3262202739715576\n",
      "\n",
      "Total adversarial train accuarcy: 48.014\n",
      "Total adversarial train loss: 531.2758046388626\n",
      "\n",
      "[ Test epoch: 23 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.76\n",
      "Current benign test loss: 0.8268281817436218\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.3845294713974\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.75\n",
      "Current benign test loss: 0.8290641903877258\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.413413405418396\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.75\n",
      "Current benign test loss: 0.8752678632736206\n",
      "Current adversarial test accuracy: 0.4\n",
      "Current adversarial test loss: 1.407863974571228\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 0.8617253303527832\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.4088401794433594\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 0.8850836753845215\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.49977707862854\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.79\n",
      "Current benign test loss: 0.8687633275985718\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.467313289642334\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 0.9369484186172485\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.4679607152938843\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 0.8378822803497314\n",
      "Current adversarial test accuracy: 0.5\n",
      "Current adversarial test loss: 1.4682893753051758\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.8631179332733154\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.4854899644851685\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.8270347118377686\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.385165810585022\n",
      "\n",
      "Total benign test accuarcy: 73.83\n",
      "Total adversarial test Accuarcy: 46.72\n",
      "Total benign test loss: 85.40076571702957\n",
      "Total adversarial test loss: 142.74955451488495\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 24 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2944684028625488\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.3111441135406494\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.3611582517623901\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.3685352802276611\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.3722783327102661\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.367871642112732\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.398561716079712\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.3008086681365967\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.398290991783142\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.1820104122161865\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.3056418895721436\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.59375\n",
      "Current adversarial train loss: 1.2250261306762695\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3910048007965088\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.5245527029037476\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3590397834777832\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2417454719543457\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.4388800859451294\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3464230298995972\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.302959680557251\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.2326730489730835\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.5117900371551514\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3493858575820923\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.3855880498886108\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.3277721405029297\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.530494213104248\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.3560152053833008\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.4099012613296509\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.2674129009246826\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3720628023147583\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2467139959335327\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.4841289520263672\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.4134511947631836\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4116170406341553\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.3275353908538818\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4265133142471313\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2333000898361206\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.3697832822799683\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.3391752243041992\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2307276725769043\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.4625\n",
      "Current adversarial train loss: 1.399505376815796\n",
      "\n",
      "Total adversarial train accuarcy: 48.492\n",
      "Total adversarial train loss: 526.434033036232\n",
      "\n",
      "[ Test epoch: 24 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.75\n",
      "Current benign test loss: 0.8274434804916382\n",
      "Current adversarial test accuracy: 0.5\n",
      "Current adversarial test loss: 1.3946439027786255\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.7906553745269775\n",
      "Current adversarial test accuracy: 0.51\n",
      "Current adversarial test loss: 1.402023434638977\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.69\n",
      "Current benign test loss: 0.8444581627845764\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.3616037368774414\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.65\n",
      "Current benign test loss: 0.9088265299797058\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.48103666305542\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.8770297765731812\n",
      "Current adversarial test accuracy: 0.52\n",
      "Current adversarial test loss: 1.4550118446350098\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.79\n",
      "Current benign test loss: 0.8091039061546326\n",
      "Current adversarial test accuracy: 0.5\n",
      "Current adversarial test loss: 1.4199414253234863\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 0.8787190914154053\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.469835638999939\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 0.8013631701469421\n",
      "Current adversarial test accuracy: 0.51\n",
      "Current adversarial test loss: 1.4345519542694092\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.79\n",
      "Current benign test loss: 0.7845112681388855\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.4003628492355347\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.75\n",
      "Current benign test loss: 0.7789325714111328\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.3334048986434937\n",
      "\n",
      "Total benign test accuarcy: 73.9\n",
      "Total adversarial test Accuarcy: 47.61\n",
      "Total benign test loss: 82.35719537734985\n",
      "Total adversarial test loss: 140.7927051782608\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 25 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.4756790399551392\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.2829216718673706\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4182208776474\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.282153844833374\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.375\n",
      "Current adversarial train loss: 1.4512830972671509\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.2407797574996948\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2545711994171143\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.3528326749801636\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.5859375\n",
      "Current adversarial train loss: 1.1722723245620728\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.3853546380996704\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.2260444164276123\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.339302897453308\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.3260917663574219\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.258060336112976\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.450212001800537\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.3820348978042603\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.4721702337265015\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.2161054611206055\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.327810525894165\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.404348373413086\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.297033667564392\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.402943730354309\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4015147686004639\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.3560792207717896\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.2570922374725342\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.243981957435608\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.2987394332885742\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.3038676977157593\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4912893772125244\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3546887636184692\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.4203530550003052\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.3089076280593872\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.1922165155410767\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.4110337495803833\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.374303936958313\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2949769496917725\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.2135560512542725\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4600484371185303\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.2561043500900269\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.5125\n",
      "Current adversarial train loss: 1.2050281763076782\n",
      "\n",
      "Total adversarial train accuarcy: 48.858\n",
      "Total adversarial train loss: 523.2777687311172\n",
      "\n",
      "[ Test epoch: 25 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.8727394342422485\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.4380220174789429\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.8358792662620544\n",
      "Current adversarial test accuracy: 0.51\n",
      "Current adversarial test loss: 1.4291033744812012\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.69\n",
      "Current benign test loss: 0.8582902550697327\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.3850210905075073\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 0.877116322517395\n",
      "Current adversarial test accuracy: 0.42\n",
      "Current adversarial test loss: 1.4192241430282593\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.88592129945755\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.4478055238723755\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.888136088848114\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.4815808534622192\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.67\n",
      "Current benign test loss: 0.9195477366447449\n",
      "Current adversarial test accuracy: 0.4\n",
      "Current adversarial test loss: 1.4694195985794067\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 0.8332111239433289\n",
      "Current adversarial test accuracy: 0.51\n",
      "Current adversarial test loss: 1.4058359861373901\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.837860643863678\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.4371566772460938\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.75\n",
      "Current benign test loss: 0.7971107363700867\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.3637455701828003\n",
      "\n",
      "Total benign test accuarcy: 71.48\n",
      "Total adversarial test Accuarcy: 46.6\n",
      "Total benign test loss: 85.31153637170792\n",
      "Total adversarial test loss: 142.08407330513\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 26 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.3424140214920044\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.4321024417877197\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.3679311275482178\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4341394901275635\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.2882766723632812\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.4775875806808472\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2568877935409546\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.193374752998352\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2477693557739258\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.2904990911483765\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.524515986442566\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3155888319015503\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.1995291709899902\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.3193870782852173\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.3385189771652222\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2724579572677612\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.4278916120529175\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.3000197410583496\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3074556589126587\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.336652159690857\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.3300963640213013\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.193092703819275\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4182606935501099\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.59375\n",
      "Current adversarial train loss: 1.1645108461380005\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.3154244422912598\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.3842192888259888\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4408155679702759\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.1838133335113525\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.333296298980713\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.3137391805648804\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.523874044418335\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4352177381515503\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.23014235496521\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4644138813018799\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3377732038497925\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.1912211179733276\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.3506752252578735\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.4810593128204346\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.566497564315796\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.3875\n",
      "Current adversarial train loss: 1.5395020246505737\n",
      "\n",
      "Total adversarial train accuarcy: 49.09\n",
      "Total adversarial train loss: 519.5412863492966\n",
      "\n",
      "[ Test epoch: 26 ]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.8676995038986206\n",
      "Current adversarial test accuracy: 0.51\n",
      "Current adversarial test loss: 1.4430744647979736\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.76\n",
      "Current benign test loss: 0.8110952973365784\n",
      "Current adversarial test accuracy: 0.58\n",
      "Current adversarial test loss: 1.4042664766311646\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.8136444687843323\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.3020384311676025\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.71\n",
      "Current benign test loss: 0.8569372296333313\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.3886488676071167\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.71\n",
      "Current benign test loss: 0.8910136222839355\n",
      "Current adversarial test accuracy: 0.5\n",
      "Current adversarial test loss: 1.4913156032562256\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.83\n",
      "Current benign test loss: 0.8707901239395142\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.4745465517044067\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 0.9051215648651123\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.4678199291229248\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.7829526662826538\n",
      "Current adversarial test accuracy: 0.5\n",
      "Current adversarial test loss: 1.3872395753860474\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.8273918032646179\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.4510005712509155\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.76\n",
      "Current benign test loss: 0.8053255677223206\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.3647030591964722\n",
      "\n",
      "Total benign test accuarcy: 75.43\n",
      "Total adversarial test Accuarcy: 47.67\n",
      "Total benign test loss: 83.43752819299698\n",
      "Total adversarial test loss: 140.43944203853607\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 27 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4731923341751099\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.28857421875\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.490634799003601\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.3270399570465088\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.235577940940857\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.252312421798706\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.59375\n",
      "Current adversarial train loss: 1.1590332984924316\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2741644382476807\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.5252693891525269\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.3498784303665161\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.277923345565796\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.539902925491333\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.4182581901550293\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.3367195129394531\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2709593772888184\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.3840276002883911\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.1929686069488525\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.274499773979187\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.2648676633834839\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.2783540487289429\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.3624423742294312\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.277063250541687\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.3800138235092163\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3222225904464722\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.23892343044281\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.59375\n",
      "Current adversarial train loss: 1.1168525218963623\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.2921080589294434\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.2574390172958374\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.4206159114837646\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.3850398063659668\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.1311421394348145\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.2879019975662231\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.3801581859588623\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.2986502647399902\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.4610294103622437\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.3323873281478882\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.2900245189666748\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.2629539966583252\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.2744613885879517\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.4875\n",
      "Current adversarial train loss: 1.3932424783706665\n",
      "\n",
      "Total adversarial train accuarcy: 49.21\n",
      "Total adversarial train loss: 514.5926332473755\n",
      "\n",
      "[ Test epoch: 27 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.76\n",
      "Current benign test loss: 0.7721547484397888\n",
      "Current adversarial test accuracy: 0.5\n",
      "Current adversarial test loss: 1.3601232767105103\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.794705867767334\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.4667346477508545\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 0.8246917724609375\n",
      "Current adversarial test accuracy: 0.42\n",
      "Current adversarial test loss: 1.3598999977111816\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.8334077596664429\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.3695753812789917\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.8493438959121704\n",
      "Current adversarial test accuracy: 0.52\n",
      "Current adversarial test loss: 1.4656760692596436\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.82\n",
      "Current benign test loss: 0.795219898223877\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.4401658773422241\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 0.879721999168396\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.4614818096160889\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.7569368481636047\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Current adversarial test accuracy: 0.53\n",
      "Current adversarial test loss: 1.3736588954925537\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.8354548811912537\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.522031545639038\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.7517316937446594\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.3239368200302124\n",
      "\n",
      "Total benign test accuarcy: 74.27\n",
      "Total adversarial test Accuarcy: 46.54\n",
      "Total benign test loss: 80.52981072664261\n",
      "Total adversarial test loss: 141.7779278755188\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 28 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.230571985244751\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.35090970993042\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.3405202627182007\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.2759722471237183\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.2900278568267822\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3051555156707764\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.3814280033111572\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3372783660888672\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2099580764770508\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.1823978424072266\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.277848482131958\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4395540952682495\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.4512759447097778\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.38908851146698\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2864657640457153\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.1583003997802734\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.231764316558838\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3197576999664307\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.3694379329681396\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.185248851776123\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.3122782707214355\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.3313419818878174\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.223974585533142\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.3381094932556152\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.3432796001434326\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.3047453165054321\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.2352945804595947\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.250204086303711\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.2590359449386597\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.322044849395752\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.258947491645813\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.286539912223816\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4621002674102783\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.3054953813552856\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.34663987159729\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3631161451339722\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2499175071716309\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2376855611801147\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.3331091403961182\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.55\n",
      "Current adversarial train loss: 1.2781472206115723\n",
      "\n",
      "Total adversarial train accuarcy: 49.698\n",
      "Total adversarial train loss: 511.53896176815033\n",
      "\n",
      "[ Test epoch: 28 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.79\n",
      "Current benign test loss: 0.8346409797668457\n",
      "Current adversarial test accuracy: 0.52\n",
      "Current adversarial test loss: 1.4282639026641846\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.75\n",
      "Current benign test loss: 0.8041954040527344\n",
      "Current adversarial test accuracy: 0.5\n",
      "Current adversarial test loss: 1.449781060218811\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 0.7872108221054077\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.3138759136199951\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.75\n",
      "Current benign test loss: 0.7954559326171875\n",
      "Current adversarial test accuracy: 0.42\n",
      "Current adversarial test loss: 1.3312568664550781\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.8193987011909485\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.4594868421554565\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.78\n",
      "Current benign test loss: 0.8583126068115234\n",
      "Current adversarial test accuracy: 0.38\n",
      "Current adversarial test loss: 1.5197439193725586\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.75\n",
      "Current benign test loss: 0.8678984045982361\n",
      "Current adversarial test accuracy: 0.38\n",
      "Current adversarial test loss: 1.4740430116653442\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.79\n",
      "Current benign test loss: 0.703514814376831\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.3188868761062622\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.76\n",
      "Current benign test loss: 0.777737557888031\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.3940136432647705\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.76\n",
      "Current benign test loss: 0.7660247683525085\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.346743106842041\n",
      "\n",
      "Total benign test accuarcy: 75.79\n",
      "Total adversarial test Accuarcy: 47.59\n",
      "Total benign test loss: 79.42464238405228\n",
      "Total adversarial test loss: 139.46258127689362\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 29 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.1461495161056519\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.4406663179397583\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.3145780563354492\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.3239179849624634\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.414469599723816\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.325488805770874\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.4186598062515259\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2533819675445557\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.425728678703308\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.406887173652649\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.2703777551651\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3654440641403198\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.176516056060791\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.4407309293746948\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.4548091888427734\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.2381272315979004\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.2431707382202148\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.2425466775894165\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.3983392715454102\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.3769030570983887\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3609479665756226\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.2338703870773315\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.4217699766159058\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.4018027782440186\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.1844993829727173\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.2841185331344604\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3769487142562866\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2226110696792603\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.321664571762085\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.157820701599121\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3419787883758545\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.1843563318252563\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.315403938293457\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.3828125\n",
      "Current adversarial train loss: 1.5368390083312988\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.2012337446212769\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2062277793884277\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2150306701660156\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4133483171463013\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.333954095840454\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.5125\n",
      "Current adversarial train loss: 1.339660406112671\n",
      "\n",
      "Total adversarial train accuarcy: 50.022\n",
      "Total adversarial train loss: 509.2930315732956\n",
      "\n",
      "[ Test epoch: 29 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.8265623450279236\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.474393367767334\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.78\n",
      "Current benign test loss: 0.755187451839447\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.3794960975646973\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.8078665137290955\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.3459129333496094\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.8593347668647766\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.4367485046386719\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.75\n",
      "Current benign test loss: 0.8694331645965576\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.5356863737106323\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.78\n",
      "Current benign test loss: 0.855312705039978\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.5192897319793701\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.71\n",
      "Current benign test loss: 0.8965410590171814\n",
      "Current adversarial test accuracy: 0.38\n",
      "Current adversarial test loss: 1.4991068840026855\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.75\n",
      "Current benign test loss: 0.765700101852417\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.4001432657241821\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.78\n",
      "Current benign test loss: 0.7568382024765015\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.39597749710083\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.7816968560218811\n",
      "Current adversarial test accuracy: 0.42\n",
      "Current adversarial test loss: 1.3775458335876465\n",
      "\n",
      "Total benign test accuarcy: 74.83\n",
      "Total adversarial test Accuarcy: 45.98\n",
      "Total benign test loss: 81.38469129800797\n",
      "Total adversarial test loss: 142.74427270889282\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 30 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3255109786987305\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3733432292938232\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.2224708795547485\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.3774843215942383\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.3590524196624756\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.2221041917800903\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.367116928100586\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3227983713150024\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.295902132987976\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.3984375\n",
      "Current adversarial train loss: 1.4116010665893555\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.1166025400161743\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.3253730535507202\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.3377779722213745\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.2738367319107056\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.4250736236572266\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.322265625\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2923574447631836\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.219216227531433\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2860190868377686\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.188539743423462\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.2863795757293701\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.360180139541626\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.3354167938232422\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.3871910572052002\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.2940095663070679\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.4908244609832764\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.2059032917022705\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.4847155809402466\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.3981127738952637\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.3011070489883423\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.1765153408050537\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.2372419834136963\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.540109395980835\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.2800618410110474\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.3749101161956787\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.285505771636963\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.3025987148284912\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.2421939373016357\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.4586601257324219\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.525\n",
      "Current adversarial train loss: 1.192901849746704\n",
      "\n",
      "Total adversarial train accuarcy: 50.34\n",
      "Total adversarial train loss: 504.9769676923752\n",
      "\n",
      "[ Test epoch: 30 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.82\n",
      "Current benign test loss: 0.8459209203720093\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.433721899986267\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.8412598967552185\n",
      "Current adversarial test accuracy: 0.55\n",
      "Current adversarial test loss: 1.4690817594528198\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 0.885571300983429\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.4078075885772705\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.75\n",
      "Current benign test loss: 0.8432442545890808\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.3590277433395386\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.8694798350334167\n",
      "Current adversarial test accuracy: 0.52\n",
      "Current adversarial test loss: 1.4783220291137695\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.79\n",
      "Current benign test loss: 0.839426577091217\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.4542583227157593\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.67\n",
      "Current benign test loss: 0.9121490716934204\n",
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.460756778717041\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.7880129218101501\n",
      "Current adversarial test accuracy: 0.53\n",
      "Current adversarial test loss: 1.4031850099563599\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.76\n",
      "Current benign test loss: 0.8018884062767029\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.4097623825073242\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 0.7995223999023438\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.3750725984573364\n",
      "\n",
      "Total benign test accuarcy: 74.88\n",
      "Total adversarial test Accuarcy: 47.37\n",
      "Total benign test loss: 83.38230097293854\n",
      "Total adversarial test loss: 140.81639897823334\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 31 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.2992185354232788\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2577065229415894\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.2980984449386597\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.2233383655548096\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.2908329963684082\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.3368586301803589\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3411277532577515\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.254227876663208\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2780532836914062\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.295937180519104\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2564756870269775\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.2256602048873901\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2467732429504395\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.2470729351043701\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.3773696422576904\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3572579622268677\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4361450672149658\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.59375\n",
      "Current adversarial train loss: 1.2163769006729126\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.1913753747940063\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.3647466897964478\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.175236701965332\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.2436113357543945\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2184451818466187\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.3415837287902832\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.1934304237365723\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.3984375\n",
      "Current adversarial train loss: 1.501747965812683\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.2250792980194092\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.2986572980880737\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.1162686347961426\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.3012465238571167\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2902727127075195\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.4625587463378906\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.2071579694747925\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.3079795837402344\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.408974528312683\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.0982680320739746\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.4634642601013184\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.2702733278274536\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.2796931266784668\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.55\n",
      "Current adversarial train loss: 1.2128952741622925\n",
      "\n",
      "Total adversarial train accuarcy: 50.27\n",
      "Total adversarial train loss: 503.2716232538223\n",
      "\n",
      "[ Test epoch: 31 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.78\n",
      "Current benign test loss: 0.8560017943382263\n",
      "Current adversarial test accuracy: 0.51\n",
      "Current adversarial test loss: 1.4442254304885864\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.71\n",
      "Current benign test loss: 0.8158104419708252\n",
      "Current adversarial test accuracy: 0.5\n",
      "Current adversarial test loss: 1.5159507989883423\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.75\n",
      "Current benign test loss: 0.8372516632080078\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.3680353164672852\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.8324920535087585\n",
      "Current adversarial test accuracy: 0.53\n",
      "Current adversarial test loss: 1.3684272766113281\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 0.9359183311462402\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.542791485786438\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.79\n",
      "Current benign test loss: 0.8476783037185669\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.476864218711853\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 0.9456693530082703\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.5047197341918945\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.788224995136261\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.3899314403533936\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.71\n",
      "Current benign test loss: 0.9112499356269836\n",
      "Current adversarial test accuracy: 0.42\n",
      "Current adversarial test loss: 1.5905506610870361\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.8114424347877502\n",
      "Current adversarial test accuracy: 0.4\n",
      "Current adversarial test loss: 1.4402644634246826\n",
      "\n",
      "Total benign test accuarcy: 72.94\n",
      "Total adversarial test Accuarcy: 46.39\n",
      "Total benign test loss: 84.40312600135803\n",
      "Total adversarial test loss: 144.80987858772278\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 32 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.2655608654022217\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2579259872436523\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.5859375\n",
      "Current adversarial train loss: 1.147569179534912\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2537846565246582\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.2459584474563599\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.217751145362854\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.3544929027557373\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.0916378498077393\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2344417572021484\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.2219980955123901\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.2808177471160889\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.3548033237457275\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.2367980480194092\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.1697208881378174\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2623233795166016\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.3551911115646362\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.3351324796676636\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.4216258525848389\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.4804766178131104\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.3650357723236084\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.301039695739746\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.2353228330612183\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.201466679573059\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.212621808052063\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2685564756393433\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.23269522190094\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2809109687805176\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.3140302896499634\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2043548822402954\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.282099962234497\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2310305833816528\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.421231985092163\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.1955689191818237\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2883059978485107\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2665907144546509\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.2788606882095337\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.3088512420654297\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4324843883514404\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.59375\n",
      "Current adversarial train loss: 1.1696863174438477\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.3782563209533691\n",
      "\n",
      "Total adversarial train accuarcy: 50.658\n",
      "Total adversarial train loss: 501.96658051013947\n",
      "\n",
      "[ Test epoch: 32 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.8\n",
      "Current benign test loss: 0.8338440656661987\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.4558411836624146\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.8403427600860596\n",
      "Current adversarial test accuracy: 0.5\n",
      "Current adversarial test loss: 1.4944980144500732\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.75\n",
      "Current benign test loss: 0.7980731129646301\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.3541449308395386\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.69\n",
      "Current benign test loss: 0.8700717091560364\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.4964371919631958\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.76\n",
      "Current benign test loss: 0.8315539360046387\n",
      "Current adversarial test accuracy: 0.5\n",
      "Current adversarial test loss: 1.4795535802841187\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.8102307319641113\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.4917351007461548\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.71\n",
      "Current benign test loss: 0.8771930932998657\n",
      "Current adversarial test accuracy: 0.4\n",
      "Current adversarial test loss: 1.4999032020568848\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.7158258557319641\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.3659826517105103\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.8064262866973877\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.4971603155136108\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.81\n",
      "Current benign test loss: 0.7207063436508179\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.3004902601242065\n",
      "\n",
      "Total benign test accuarcy: 76.15\n",
      "Total adversarial test Accuarcy: 46.91\n",
      "Total benign test loss: 79.75611931085587\n",
      "Total adversarial test loss: 142.2260365486145\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 33 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.310191035270691\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.59375\n",
      "Current adversarial train loss: 1.0536898374557495\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2644940614700317\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2889490127563477\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3489326238632202\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.15185546875\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.340665578842163\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3260152339935303\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2798672914505005\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.1855840682983398\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.1888885498046875\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.331304907798767\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.1656818389892578\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3443293571472168\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.609375\n",
      "Current adversarial train loss: 1.1298445463180542\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.197490930557251\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.2490071058273315\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.2651216983795166\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.1612889766693115\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2871320247650146\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2666631937026978\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.2316843271255493\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2659705877304077\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.322605013847351\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.3897721767425537\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.3090537786483765\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.2970727682113647\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.390625\n",
      "Current adversarial train loss: 1.4667892456054688\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3882416486740112\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.235676646232605\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2101221084594727\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.151842474937439\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.471729040145874\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.2688912153244019\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2417948246002197\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.2906758785247803\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.2930296659469604\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2188005447387695\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3124407529830933\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.4875\n",
      "Current adversarial train loss: 1.2168744802474976\n",
      "\n",
      "Total adversarial train accuarcy: 50.892\n",
      "Total adversarial train loss: 498.4935528039932\n",
      "\n",
      "[ Test epoch: 33 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.81\n",
      "Current benign test loss: 0.7689972519874573\n",
      "Current adversarial test accuracy: 0.54\n",
      "Current adversarial test loss: 1.3654570579528809\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.75\n",
      "Current benign test loss: 0.7754070162773132\n",
      "Current adversarial test accuracy: 0.51\n",
      "Current adversarial test loss: 1.4428669214248657\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 0.8068163394927979\n",
      "Current adversarial test accuracy: 0.36\n",
      "Current adversarial test loss: 1.374192714691162\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.76\n",
      "Current benign test loss: 0.8040776252746582\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.3421330451965332\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.75\n",
      "Current benign test loss: 0.8177106380462646\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.4747394323349\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.86\n",
      "Current benign test loss: 0.815177321434021\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.4568225145339966\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.8451537489891052\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.4206740856170654\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.8\n",
      "Current benign test loss: 0.7132830023765564\n",
      "Current adversarial test accuracy: 0.5\n",
      "Current adversarial test loss: 1.3465150594711304\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.7922257781028748\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.4284758567810059\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.82\n",
      "Current benign test loss: 0.6943652629852295\n",
      "Current adversarial test accuracy: 0.52\n",
      "Current adversarial test loss: 1.235957384109497\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Total benign test accuarcy: 77.1\n",
      "Total adversarial test Accuarcy: 47.72\n",
      "Total benign test loss: 77.50985038280487\n",
      "Total adversarial test loss: 137.86439681053162\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 34 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.3311313390731812\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2341986894607544\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.0901869535446167\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.2283110618591309\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2874561548233032\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.2661099433898926\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.6640625\n",
      "Current adversarial train loss: 1.0287959575653076\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.1733142137527466\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.1984679698944092\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.1597964763641357\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2787179946899414\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.1235580444335938\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2351974248886108\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.219028353691101\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.231927514076233\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.3002535104751587\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.4097336530685425\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3054015636444092\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3172461986541748\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3479772806167603\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3537980318069458\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2043828964233398\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.327268362045288\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.1726102828979492\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.243666172027588\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.304864525794983\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.1910548210144043\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.1537880897521973\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.4812711477279663\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.251692295074463\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.4226394891738892\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.352899193763733\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2160358428955078\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.402558445930481\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3210809230804443\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.1517183780670166\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2132415771484375\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.3676424026489258\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.2967320680618286\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.425\n",
      "Current adversarial train loss: 1.3966619968414307\n",
      "\n",
      "Total adversarial train accuarcy: 51.146\n",
      "Total adversarial train loss: 494.8570740222931\n",
      "\n",
      "[ Test epoch: 34 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.8\n",
      "Current benign test loss: 0.7717748284339905\n",
      "Current adversarial test accuracy: 0.5\n",
      "Current adversarial test loss: 1.3700649738311768\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.7649390697479248\n",
      "Current adversarial test accuracy: 0.52\n",
      "Current adversarial test loss: 1.4301620721817017\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 0.8528593182563782\n",
      "Current adversarial test accuracy: 0.38\n",
      "Current adversarial test loss: 1.413820743560791\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 0.8436813950538635\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.4291459321975708\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.867665708065033\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.5269272327423096\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.8479278683662415\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.5196335315704346\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.67\n",
      "Current benign test loss: 0.9217414259910583\n",
      "Current adversarial test accuracy: 0.38\n",
      "Current adversarial test loss: 1.5407838821411133\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.69\n",
      "Current benign test loss: 0.7351261973381042\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.3741382360458374\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.75\n",
      "Current benign test loss: 0.7843097448348999\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.4315801858901978\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.75\n",
      "Current benign test loss: 0.7402722239494324\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.337165355682373\n",
      "\n",
      "Total benign test accuarcy: 73.54\n",
      "Total adversarial test Accuarcy: 47.21\n",
      "Total benign test loss: 79.95336174964905\n",
      "Total adversarial test loss: 141.1136450767517\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 35 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.3210653066635132\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.1943992376327515\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.2742486000061035\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.286089301109314\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.150675892829895\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.3102673292160034\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.4237042665481567\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2322819232940674\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.4387272596359253\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.2186344861984253\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.1621612310409546\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.1733309030532837\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.28065025806427\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.2487258911132812\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2299277782440186\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.304533839225769\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3908836841583252\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.2361938953399658\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.3241195678710938\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.1987067461013794\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3512309789657593\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.2303754091262817\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.3673604726791382\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.2823208570480347\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.59375\n",
      "Current adversarial train loss: 1.1021825075149536\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.1580398082733154\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.3874183893203735\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.278772234916687\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.2066725492477417\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2369364500045776\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.2452532052993774\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.1241823434829712\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.324812650680542\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.2639213800430298\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.2962297201156616\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.2672035694122314\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2292816638946533\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.3006278276443481\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.2087335586547852\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.4875\n",
      "Current adversarial train loss: 1.356895089149475\n",
      "\n",
      "Total adversarial train accuarcy: 51.152\n",
      "Total adversarial train loss: 492.0851916074753\n",
      "\n",
      "[ Test epoch: 35 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.79\n",
      "Current benign test loss: 0.8136951923370361\n",
      "Current adversarial test accuracy: 0.53\n",
      "Current adversarial test loss: 1.3699191808700562\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.7806060910224915\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.428725242614746\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.64\n",
      "Current benign test loss: 0.8570142388343811\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.3971896171569824\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.63\n",
      "Current benign test loss: 0.8665218949317932\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.4171127080917358\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.64\n",
      "Current benign test loss: 0.8806741237640381\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.4782943725585938\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.78\n",
      "Current benign test loss: 0.8230811953544617\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.445518970489502\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.64\n",
      "Current benign test loss: 0.9515842199325562\n",
      "Current adversarial test accuracy: 0.34\n",
      "Current adversarial test loss: 1.5484176874160767\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.71\n",
      "Current benign test loss: 0.7497237920761108\n",
      "Current adversarial test accuracy: 0.52\n",
      "Current adversarial test loss: 1.3573700189590454\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.7873169779777527\n",
      "Current adversarial test accuracy: 0.52\n",
      "Current adversarial test loss: 1.4090385437011719\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.71\n",
      "Current benign test loss: 0.7731784582138062\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.358831763267517\n",
      "\n",
      "Total benign test accuarcy: 72.43\n",
      "Total adversarial test Accuarcy: 47.86\n",
      "Total benign test loss: 82.29362833499908\n",
      "Total adversarial test loss: 140.62117993831635\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 36 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.180090069770813\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.1732197999954224\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.473990559577942\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.3635382652282715\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2714016437530518\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.3080551624298096\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.239419937133789\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.329525113105774\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.6171875\n",
      "Current adversarial train loss: 1.1234033107757568\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2726342678070068\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.1640805006027222\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.1128813028335571\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.200373888015747\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.328439712524414\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3062366247177124\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3674362897872925\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.1969122886657715\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4363600015640259\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2150520086288452\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2162524461746216\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.1883668899536133\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.2812987565994263\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.4185775518417358\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.1393033266067505\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.262717843055725\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.3482493162155151\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.213017463684082\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2833021879196167\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.2891536951065063\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.2672675848007202\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2268449068069458\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.3027949333190918\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4866352081298828\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.30604887008667\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.1759995222091675\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.242296814918518\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.305309534072876\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2307449579238892\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.2007116079330444\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.5125\n",
      "Current adversarial train loss: 1.2148699760437012\n",
      "\n",
      "Total adversarial train accuarcy: 51.418\n",
      "Total adversarial train loss: 490.7719863653183\n",
      "\n",
      "[ Test epoch: 36 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.81\n",
      "Current benign test loss: 0.8093547224998474\n",
      "Current adversarial test accuracy: 0.54\n",
      "Current adversarial test loss: 1.4357993602752686\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.7142742276191711\n",
      "Current adversarial test accuracy: 0.57\n",
      "Current adversarial test loss: 1.3566501140594482\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.8346010446548462\n",
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.3949699401855469\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.71\n",
      "Current benign test loss: 0.8793231844902039\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.4748077392578125\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.8311378359794617\n",
      "Current adversarial test accuracy: 0.51\n",
      "Current adversarial test loss: 1.497145414352417\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.81\n",
      "Current benign test loss: 0.7861219644546509\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.4548989534378052\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.8318171501159668\n",
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.4334769248962402\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.75\n",
      "Current benign test loss: 0.7008755207061768\n",
      "Current adversarial test accuracy: 0.5\n",
      "Current adversarial test loss: 1.3538918495178223\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.76\n",
      "Current benign test loss: 0.7656621336936951\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.4463887214660645\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.8\n",
      "Current benign test loss: 0.7162911891937256\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.31276273727417\n",
      "\n",
      "Total benign test accuarcy: 76.26\n",
      "Total adversarial test Accuarcy: 48.25\n",
      "Total benign test loss: 77.78483873605728\n",
      "Total adversarial test loss: 138.78374314308167\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 37 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.126307487487793\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.2185224294662476\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.2440580129623413\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2001632452011108\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.1257706880569458\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.3394427299499512\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.5859375\n",
      "Current adversarial train loss: 1.2101141214370728\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.2994585037231445\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.2924392223358154\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2826799154281616\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2857356071472168\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.1821540594100952\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.2807464599609375\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2510147094726562\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.625\n",
      "Current adversarial train loss: 1.0934109687805176\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.358620524406433\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.3049741983413696\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.3671985864639282\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2201588153839111\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.202885627746582\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.3083593845367432\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.3192596435546875\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.178520679473877\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2323546409606934\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.2679985761642456\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.6015625\n",
      "Current adversarial train loss: 1.1250362396240234\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.59375\n",
      "Current adversarial train loss: 1.1770586967468262\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.2245444059371948\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.1984140872955322\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2452669143676758\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.3031048774719238\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.2156362533569336\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.2422059774398804\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.208206295967102\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.3545947074890137\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3035743236541748\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.137043833732605\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.2405608892440796\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.2878249883651733\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.45\n",
      "Current adversarial train loss: 1.4404776096343994\n",
      "\n",
      "Total adversarial train accuarcy: 51.742\n",
      "Total adversarial train loss: 488.66826379299164\n",
      "\n",
      "[ Test epoch: 37 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.7508310079574585\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.4205045700073242\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.78\n",
      "Current benign test loss: 0.7008934020996094\n",
      "Current adversarial test accuracy: 0.53\n",
      "Current adversarial test loss: 1.4025092124938965\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.772063136100769\n",
      "Current adversarial test accuracy: 0.38\n",
      "Current adversarial test loss: 1.3918887376785278\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.76\n",
      "Current benign test loss: 0.7864043712615967\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.4431103467941284\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.79\n",
      "Current benign test loss: 0.712256669998169\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.3809036016464233\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.84\n",
      "Current benign test loss: 0.7098813652992249\n",
      "Current adversarial test accuracy: 0.51\n",
      "Current adversarial test loss: 1.3999780416488647\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.8227920532226562\n",
      "Current adversarial test accuracy: 0.38\n",
      "Current adversarial test loss: 1.517823338508606\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.76\n",
      "Current benign test loss: 0.733501672744751\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.4717075824737549\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.81\n",
      "Current benign test loss: 0.692950427532196\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.3577053546905518\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.7448400259017944\n",
      "Current adversarial test accuracy: 0.42\n",
      "Current adversarial test loss: 1.4440231323242188\n",
      "\n",
      "Total benign test accuarcy: 77.71\n",
      "Total adversarial test Accuarcy: 47.36\n",
      "Total benign test loss: 73.11851507425308\n",
      "Total adversarial test loss: 140.4052402973175\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 38 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3981115818023682\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.6328125\n",
      "Current adversarial train loss: 1.0642974376678467\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.2683221101760864\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.26376211643219\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.1962662935256958\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.2158244848251343\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.2764708995819092\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2363089323043823\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.2940897941589355\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.2920054197311401\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.1550383567810059\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.4011954069137573\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.1478854417800903\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.2477457523345947\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.2238863706588745\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.2185841798782349\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3448748588562012\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.2176209688186646\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.59375\n",
      "Current adversarial train loss: 1.109628438949585\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.3207147121429443\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2332866191864014\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.252623438835144\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.3376836776733398\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.2419415712356567\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.2534037828445435\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.2303926944732666\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2119437456130981\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.2408905029296875\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.2635905742645264\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.259556531906128\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.1948524713516235\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.1942609548568726\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.172678828239441\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.2469861507415771\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.1559442281723022\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.4787046909332275\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.609375\n",
      "Current adversarial train loss: 1.13071608543396\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.2668945789337158\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.2707704305648804\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.2523006200790405\n",
      "\n",
      "Total adversarial train accuarcy: 51.862\n",
      "Total adversarial train loss: 484.5073212981224\n",
      "\n",
      "[ Test epoch: 38 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.79\n",
      "Current benign test loss: 0.8011607527732849\n",
      "Current adversarial test accuracy: 0.5\n",
      "Current adversarial test loss: 1.4262386560440063\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.78\n",
      "Current benign test loss: 0.7268686890602112\n",
      "Current adversarial test accuracy: 0.54\n",
      "Current adversarial test loss: 1.3872795104980469\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.7830679416656494\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.307032823562622\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.8309135437011719\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.415716528892517\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.78\n",
      "Current benign test loss: 0.8307758569717407\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Current adversarial test accuracy: 0.52\n",
      "Current adversarial test loss: 1.4647413492202759\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.81\n",
      "Current benign test loss: 0.7992578148841858\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.5030256509780884\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 0.8946201801300049\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.4743353128433228\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.79\n",
      "Current benign test loss: 0.7015950083732605\n",
      "Current adversarial test accuracy: 0.5\n",
      "Current adversarial test loss: 1.3167784214019775\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.82\n",
      "Current benign test loss: 0.7313123941421509\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.3617563247680664\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.7404431700706482\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.3581293821334839\n",
      "\n",
      "Total benign test accuarcy: 77.32\n",
      "Total adversarial test Accuarcy: 49.55\n",
      "Total benign test loss: 77.17986488342285\n",
      "Total adversarial test loss: 137.58102548122406\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 39 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.6015625\n",
      "Current adversarial train loss: 1.1497502326965332\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.2040939331054688\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3043633699417114\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2538905143737793\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.6015625\n",
      "Current adversarial train loss: 1.1167945861816406\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.1320170164108276\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.2368260622024536\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.5859375\n",
      "Current adversarial train loss: 1.099990725517273\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.1595159769058228\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.0632421970367432\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2090193033218384\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.4296875\n",
      "Current adversarial train loss: 1.3362921476364136\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.1436065435409546\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.204094648361206\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3468912839889526\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.5859375\n",
      "Current adversarial train loss: 1.1329118013381958\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.176499366760254\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.1850076913833618\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3521767854690552\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.209826111793518\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.1804717779159546\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.2345134019851685\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.1765832901000977\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.5859375\n",
      "Current adversarial train loss: 1.1103546619415283\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.216218113899231\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.1844276189804077\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.273547649383545\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.2006245851516724\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3262441158294678\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.2839716672897339\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.1467781066894531\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2369579076766968\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.140012264251709\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.287765622138977\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.1940594911575317\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.1806857585906982\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.2910442352294922\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.2237147092819214\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2646678686141968\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.3875\n",
      "Current adversarial train loss: 1.4590933322906494\n",
      "\n",
      "Total adversarial train accuarcy: 51.99\n",
      "Total adversarial train loss: 484.6284593343735\n",
      "\n",
      "[ Test epoch: 39 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.83\n",
      "Current benign test loss: 0.7348825335502625\n",
      "Current adversarial test accuracy: 0.51\n",
      "Current adversarial test loss: 1.355981707572937\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.78\n",
      "Current benign test loss: 0.7443300485610962\n",
      "Current adversarial test accuracy: 0.53\n",
      "Current adversarial test loss: 1.4084107875823975\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.7844284176826477\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.3366116285324097\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.7918963432312012\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.4051088094711304\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.75\n",
      "Current benign test loss: 0.7817608118057251\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.4502308368682861\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.79\n",
      "Current benign test loss: 0.7802597880363464\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.4422465562820435\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 0.901702880859375\n",
      "Current adversarial test accuracy: 0.4\n",
      "Current adversarial test loss: 1.5354931354522705\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.6920140981674194\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.3207718133926392\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.81\n",
      "Current benign test loss: 0.7610774040222168\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.4660406112670898\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.75\n",
      "Current benign test loss: 0.731558620929718\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.391525149345398\n",
      "\n",
      "Total benign test accuarcy: 77.2\n",
      "Total adversarial test Accuarcy: 48.04\n",
      "Total benign test loss: 75.9327774643898\n",
      "Total adversarial test loss: 139.6294457912445\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 40 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.2314950227737427\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.228318452835083\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.59375\n",
      "Current adversarial train loss: 1.0900914669036865\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.399423360824585\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.111358404159546\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.165576696395874\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.13298761844635\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.1085224151611328\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.201219081878662\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.1168614625930786\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.1776689291000366\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.172764778137207\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.2959786653518677\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.0500186681747437\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.2466933727264404\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.3021571636199951\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.0731168985366821\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.3141300678253174\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.2431443929672241\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.2240670919418335\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2982118129730225\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.225135326385498\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.1567734479904175\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2771012783050537\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.1644299030303955\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.59375\n",
      "Current adversarial train loss: 1.1937986612319946\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2962921857833862\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.2898467779159546\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.1464394330978394\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.3924615383148193\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.4250032901763916\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2455780506134033\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.3849986791610718\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.3756860494613647\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.3519359827041626\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.3118839263916016\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2739170789718628\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.3061109781265259\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3217592239379883\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.475\n",
      "Current adversarial train loss: 1.35488760471344\n",
      "\n",
      "Total adversarial train accuarcy: 52.202\n",
      "Total adversarial train loss: 480.6717250943184\n",
      "\n",
      "[ Test epoch: 40 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.7896420359611511\n",
      "Current adversarial test accuracy: 0.5\n",
      "Current adversarial test loss: 1.4268097877502441\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.7713549137115479\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.442923903465271\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.7741855382919312\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.339410662651062\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 0.8791370987892151\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.477720022201538\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.69\n",
      "Current benign test loss: 0.8103367686271667\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.438761591911316\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.78\n",
      "Current benign test loss: 0.8239383101463318\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.5128757953643799\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.64\n",
      "Current benign test loss: 0.8872260451316833\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.5078520774841309\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.7128410935401917\n",
      "Current adversarial test accuracy: 0.57\n",
      "Current adversarial test loss: 1.344981074333191\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.79\n",
      "Current benign test loss: 0.7893069386482239\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.4677307605743408\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.8\n",
      "Current benign test loss: 0.6847048401832581\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.300710916519165\n",
      "\n",
      "Total benign test accuarcy: 74.28\n",
      "Total adversarial test Accuarcy: 47.45\n",
      "Total benign test loss: 78.07653975486755\n",
      "Total adversarial test loss: 140.12986743450165\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 41 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.2821929454803467\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.1153644323349\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.609375\n",
      "Current adversarial train loss: 1.060575008392334\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.2674378156661987\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.3701878786087036\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2667932510375977\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3373793363571167\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.6171875\n",
      "Current adversarial train loss: 1.1071038246154785\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.1557008028030396\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.59375\n",
      "Current adversarial train loss: 1.2032045125961304\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2823573350906372\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.283806324005127\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.3126661777496338\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2548295259475708\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2458864450454712\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.1719146966934204\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.1900712251663208\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2377246618270874\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.1676033735275269\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.130181908607483\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.170080542564392\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.1357067823410034\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.1808170080184937\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.2991421222686768\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.366986870765686\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.1959712505340576\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.239932656288147\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.2320584058761597\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.2437267303466797\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2907354831695557\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.1225666999816895\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.1776214838027954\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.2890598773956299\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.310787558555603\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.166851282119751\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.2374818325042725\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.1733022928237915\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2501466274261475\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3028204441070557\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.6\n",
      "Current adversarial train loss: 1.0665594339370728\n",
      "\n",
      "Total adversarial train accuarcy: 52.662\n",
      "Total adversarial train loss: 478.41386020183563\n",
      "\n",
      "[ Test epoch: 41 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.71\n",
      "Current benign test loss: 0.896382749080658\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.5038169622421265\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.75\n",
      "Current benign test loss: 0.8450415730476379\n",
      "Current adversarial test accuracy: 0.51\n",
      "Current adversarial test loss: 1.5169575214385986\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.69\n",
      "Current benign test loss: 0.829497218132019\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.3549940586090088\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.8096899390220642\n",
      "Current adversarial test accuracy: 0.53\n",
      "Current adversarial test loss: 1.3406916856765747\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.78\n",
      "Current benign test loss: 0.7785304188728333\n",
      "Current adversarial test accuracy: 0.51\n",
      "Current adversarial test loss: 1.3798176050186157\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.83\n",
      "Current benign test loss: 0.7830897569656372\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.4395381212234497\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 0.8823506832122803\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.4775903224945068\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.7853541374206543\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.4158517122268677\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.81\n",
      "Current benign test loss: 0.7289275527000427\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.3436561822891235\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.76\n",
      "Current benign test loss: 0.7687407732009888\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.363420009613037\n",
      "\n",
      "Total benign test accuarcy: 76.04\n",
      "Total adversarial test Accuarcy: 49.24\n",
      "Total benign test loss: 78.62771707773209\n",
      "Total adversarial test loss: 137.67013776302338\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 42 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.1745859384536743\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2473958730697632\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.1563838720321655\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.2462894916534424\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2180715799331665\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.1976902484893799\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2797942161560059\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.2457301616668701\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.259550929069519\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.59375\n",
      "Current adversarial train loss: 1.1704437732696533\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.2049903869628906\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.1348835229873657\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2721794843673706\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.2932014465332031\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.59375\n",
      "Current adversarial train loss: 1.139699101448059\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.1768978834152222\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.2185851335525513\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.2245848178863525\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.2854810953140259\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.223271131515503\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2712771892547607\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2620927095413208\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.4562911987304688\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.175171136856079\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.129163384437561\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.1880954504013062\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.6171875\n",
      "Current adversarial train loss: 1.0810328722000122\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3592110872268677\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.3085064888000488\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.1454311609268188\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.189693808555603\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.215920090675354\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2833952903747559\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.316576361656189\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.1752285957336426\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.337196946144104\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2841826677322388\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.1976003646850586\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3076262474060059\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.3584072589874268\n",
      "\n",
      "Total adversarial train accuarcy: 52.502\n",
      "Total adversarial train loss: 477.1269145011902\n",
      "\n",
      "[ Test epoch: 42 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.81\n",
      "Current benign test loss: 0.7501400709152222\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.39412260055542\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.76\n",
      "Current benign test loss: 0.7069214582443237\n",
      "Current adversarial test accuracy: 0.51\n",
      "Current adversarial test loss: 1.344659447669983\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.76\n",
      "Current benign test loss: 0.7416848540306091\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.3010557889938354\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.8080499768257141\n",
      "Current adversarial test accuracy: 0.5\n",
      "Current adversarial test loss: 1.4111394882202148\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.7521547675132751\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.4047552347183228\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.85\n",
      "Current benign test loss: 0.7287771105766296\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.407011866569519\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 0.8290092349052429\n",
      "Current adversarial test accuracy: 0.4\n",
      "Current adversarial test loss: 1.4589191675186157\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.75\n",
      "Current benign test loss: 0.6907713413238525\n",
      "Current adversarial test accuracy: 0.51\n",
      "Current adversarial test loss: 1.3668535947799683\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.81\n",
      "Current benign test loss: 0.7372586131095886\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.3853142261505127\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.79\n",
      "Current benign test loss: 0.7129048109054565\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.3045955896377563\n",
      "\n",
      "Total benign test accuarcy: 77.05\n",
      "Total adversarial test Accuarcy: 48.08\n",
      "Total benign test loss: 74.32448780536652\n",
      "Total adversarial test loss: 136.69206142425537\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 43 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.09225332736969\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.1762282848358154\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.213278889656067\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.2579878568649292\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2241886854171753\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2456213235855103\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.1867666244506836\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.2037700414657593\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.1316967010498047\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2852857112884521\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.2274792194366455\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.2712470293045044\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2664966583251953\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.3075778484344482\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.065149188041687\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.3228448629379272\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.158405065536499\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.171004056930542\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.3791650533676147\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.2438766956329346\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.65625\n",
      "Current adversarial train loss: 1.0044474601745605\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.2477023601531982\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2595640420913696\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.393945336341858\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.1549814939498901\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.1328855752944946\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.1652766466140747\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2436206340789795\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.190556526184082\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.202785849571228\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.2101380825042725\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3415100574493408\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.2658919095993042\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.1543196439743042\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.2204371690750122\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.6015625\n",
      "Current adversarial train loss: 1.0697463750839233\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2503257989883423\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.1993480920791626\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2038459777832031\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.55\n",
      "Current adversarial train loss: 1.2171986103057861\n",
      "\n",
      "Total adversarial train accuarcy: 52.696\n",
      "Total adversarial train loss: 474.2152283191681\n",
      "\n",
      "[ Test epoch: 43 ]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.82\n",
      "Current benign test loss: 0.736713707447052\n",
      "Current adversarial test accuracy: 0.52\n",
      "Current adversarial test loss: 1.3777917623519897\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.7437928915023804\n",
      "Current adversarial test accuracy: 0.51\n",
      "Current adversarial test loss: 1.4535330533981323\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.7467833161354065\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.3201855421066284\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.7960160970687866\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.4611620903015137\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.7131184935569763\n",
      "Current adversarial test accuracy: 0.51\n",
      "Current adversarial test loss: 1.411169409751892\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.85\n",
      "Current benign test loss: 0.6846105456352234\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.417152762413025\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.8404707312583923\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.5071063041687012\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.78\n",
      "Current benign test loss: 0.6487173438072205\n",
      "Current adversarial test accuracy: 0.5\n",
      "Current adversarial test loss: 1.3009861707687378\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.81\n",
      "Current benign test loss: 0.6977372169494629\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.3661004304885864\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.7290608286857605\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.3813116550445557\n",
      "\n",
      "Total benign test accuarcy: 77.26\n",
      "Total adversarial test Accuarcy: 48.0\n",
      "Total benign test loss: 72.68553060293198\n",
      "Total adversarial test loss: 139.40610909461975\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 44 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.40625\n",
      "Current adversarial train loss: 1.3523244857788086\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.59375\n",
      "Current adversarial train loss: 1.1332247257232666\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.5859375\n",
      "Current adversarial train loss: 1.0902106761932373\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.154718041419983\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2171721458435059\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.2512907981872559\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2626805305480957\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.061532974243164\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.2276005744934082\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.6015625\n",
      "Current adversarial train loss: 1.2025079727172852\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.0630913972854614\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3381493091583252\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2172449827194214\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.223958134651184\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.3233202695846558\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.59375\n",
      "Current adversarial train loss: 1.0778595209121704\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.1211316585540771\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.4042766094207764\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.1598782539367676\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.1398013830184937\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.1706762313842773\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.2768526077270508\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2769849300384521\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.2700169086456299\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.379412293434143\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.2393152713775635\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2731667757034302\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.2196986675262451\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.209008812904358\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.1213929653167725\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.0334962606430054\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.1353691816329956\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.59375\n",
      "Current adversarial train loss: 1.0478012561798096\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.1627713441848755\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2693512439727783\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.4375\n",
      "Current adversarial train loss: 1.3318108320236206\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.1400197744369507\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.3425090312957764\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.1592971086502075\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.4875\n",
      "Current adversarial train loss: 1.1972872018814087\n",
      "\n",
      "Total adversarial train accuarcy: 53.04\n",
      "Total adversarial train loss: 470.80256628990173\n",
      "\n",
      "[ Test epoch: 44 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.8056724071502686\n",
      "Current adversarial test accuracy: 0.51\n",
      "Current adversarial test loss: 1.444590449333191\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.8\n",
      "Current benign test loss: 0.7496147751808167\n",
      "Current adversarial test accuracy: 0.5\n",
      "Current adversarial test loss: 1.401822805404663\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 0.8457962870597839\n",
      "Current adversarial test accuracy: 0.4\n",
      "Current adversarial test loss: 1.4498367309570312\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.8331124186515808\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.448055386543274\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.7790551781654358\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.4276199340820312\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.85\n",
      "Current benign test loss: 0.7969698905944824\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.5263657569885254\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.8594439029693604\n",
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.479494333267212\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.75\n",
      "Current benign test loss: 0.7039729952812195\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Current adversarial test accuracy: 0.51\n",
      "Current adversarial test loss: 1.3687825202941895\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.81\n",
      "Current benign test loss: 0.7565192580223083\n",
      "Current adversarial test accuracy: 0.49\n",
      "Current adversarial test loss: 1.4207255840301514\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.7093693614006042\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.31264066696167\n",
      "\n",
      "Total benign test accuarcy: 75.62\n",
      "Total adversarial test Accuarcy: 47.63\n",
      "Total benign test loss: 77.93828374147415\n",
      "Total adversarial test loss: 141.50200772285461\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 45 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.4246103763580322\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.1903305053710938\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2550697326660156\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.6015625\n",
      "Current adversarial train loss: 1.1304569244384766\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.162685751914978\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.2096258401870728\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.1948747634887695\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.222447156906128\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.1131473779678345\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2529735565185547\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.0499951839447021\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.232007384300232\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.5859375\n",
      "Current adversarial train loss: 1.0964906215667725\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2533470392227173\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2687031030654907\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.6015625\n",
      "Current adversarial train loss: 1.1684693098068237\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.1952253580093384\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.6015625\n",
      "Current adversarial train loss: 1.105650544166565\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.3516736030578613\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.2030853033065796\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.3804224729537964\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.5859375\n",
      "Current adversarial train loss: 1.113028883934021\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.347660779953003\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.1759076118469238\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.3488292694091797\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.2223197221755981\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2814072370529175\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.59375\n",
      "Current adversarial train loss: 1.0167328119277954\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3532774448394775\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2256081104278564\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.2204790115356445\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2355730533599854\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.1853382587432861\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.10903799533844\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.1640689373016357\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.1514921188354492\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.2299392223358154\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.4166215658187866\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.2144991159439087\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.325\n",
      "Current adversarial train loss: 1.4567244052886963\n",
      "\n",
      "Total adversarial train accuarcy: 53.022\n",
      "Total adversarial train loss: 471.6432752609253\n",
      "\n",
      "[ Test epoch: 45 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.78\n",
      "Current benign test loss: 0.7679151892662048\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.4715111255645752\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.78\n",
      "Current benign test loss: 0.7287666201591492\n",
      "Current adversarial test accuracy: 0.51\n",
      "Current adversarial test loss: 1.4406994581222534\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.78\n",
      "Current benign test loss: 0.7469473481178284\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.3514609336853027\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.7621010541915894\n",
      "Current adversarial test accuracy: 0.51\n",
      "Current adversarial test loss: 1.3893779516220093\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.762937605381012\n",
      "Current adversarial test accuracy: 0.42\n",
      "Current adversarial test loss: 1.4667232036590576\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.87\n",
      "Current benign test loss: 0.7140292525291443\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.433824062347412\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.76\n",
      "Current benign test loss: 0.7918376922607422\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.4738094806671143\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.6805378794670105\n",
      "Current adversarial test accuracy: 0.51\n",
      "Current adversarial test loss: 1.3547613620758057\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.83\n",
      "Current benign test loss: 0.6834897398948669\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.3575835227966309\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.79\n",
      "Current benign test loss: 0.6911672353744507\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.3369320631027222\n",
      "\n",
      "Total benign test accuarcy: 78.68\n",
      "Total adversarial test Accuarcy: 48.89\n",
      "Total benign test loss: 70.66505932807922\n",
      "Total adversarial test loss: 136.52493858337402\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 46 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.5859375\n",
      "Current adversarial train loss: 1.1128443479537964\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.59375\n",
      "Current adversarial train loss: 1.01846182346344\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.28842294216156\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.6328125\n",
      "Current adversarial train loss: 1.0614722967147827\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.5859375\n",
      "Current adversarial train loss: 1.0815547704696655\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.1732594966888428\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2585015296936035\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.2042567729949951\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.1971871852874756\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.135561227798462\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.3648004531860352\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.2266921997070312\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.1203465461730957\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.2459228038787842\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.3234546184539795\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.1175366640090942\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.5859375\n",
      "Current adversarial train loss: 1.1953442096710205\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.1031490564346313\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.2165721654891968\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.1273329257965088\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.4304755926132202\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.307431697845459\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.2791295051574707\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2576361894607544\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.0708420276641846\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.2883317470550537\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.1340543031692505\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.2197717428207397\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.2273651361465454\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.6015625\n",
      "Current adversarial train loss: 1.091390609741211\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.1231060028076172\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.4140625\n",
      "Current adversarial train loss: 1.4228464365005493\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.1693360805511475\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.2556008100509644\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.3231521844863892\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.2755584716796875\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2972545623779297\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.1985468864440918\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.214546799659729\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.5125\n",
      "Current adversarial train loss: 1.148545265197754\n",
      "\n",
      "Total adversarial train accuarcy: 52.946\n",
      "Total adversarial train loss: 470.01350259780884\n",
      "\n",
      "[ Test epoch: 46 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.78\n",
      "Current benign test loss: 0.8112501502037048\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.4555391073226929\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.7007275223731995\n",
      "Current adversarial test accuracy: 0.54\n",
      "Current adversarial test loss: 1.3245110511779785\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.8004884123802185\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.3508367538452148\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.69\n",
      "Current benign test loss: 0.8475740551948547\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.4482365846633911\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.8114847540855408\n",
      "Current adversarial test accuracy: 0.55\n",
      "Current adversarial test loss: 1.455793857574463\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.84\n",
      "Current benign test loss: 0.7835540175437927\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.4673068523406982\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 0.8855779767036438\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.520674705505371\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.7023527026176453\n",
      "Current adversarial test accuracy: 0.54\n",
      "Current adversarial test loss: 1.310922384262085\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.79\n",
      "Current benign test loss: 0.7271436452865601\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.3679105043411255\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.8\n",
      "Current benign test loss: 0.7076742053031921\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.2937544584274292\n",
      "\n",
      "Total benign test accuarcy: 75.74\n",
      "Total adversarial test Accuarcy: 48.66\n",
      "Total benign test loss: 76.27114909887314\n",
      "Total adversarial test loss: 137.74259638786316\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 47 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.5859375\n",
      "Current adversarial train loss: 1.1246119737625122\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.177963137626648\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2991093397140503\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.3506337404251099\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.14735746383667\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.236650824546814\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.3094666004180908\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.3047146797180176\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.262333869934082\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.3207303285598755\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.098822832107544\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.314573884010315\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.1023024320602417\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.6015625\n",
      "Current adversarial train loss: 1.0970513820648193\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.5859375\n",
      "Current adversarial train loss: 1.1670756340026855\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.6015625\n",
      "Current adversarial train loss: 1.1496202945709229\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.0553147792816162\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.1249268054962158\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.1066449880599976\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.59375\n",
      "Current adversarial train loss: 1.171142816543579\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3337972164154053\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.4765625\n",
      "Current adversarial train loss: 1.281655192375183\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.3118432760238647\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.3203539848327637\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.1814725399017334\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.1344530582427979\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2296943664550781\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2087095975875854\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.2313227653503418\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.1833972930908203\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.609375\n",
      "Current adversarial train loss: 1.1393744945526123\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.625\n",
      "Current adversarial train loss: 1.1235674619674683\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.220818042755127\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.1238255500793457\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.1811676025390625\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.1904371976852417\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.1866512298583984\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2789795398712158\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.453125\n",
      "Current adversarial train loss: 1.309922695159912\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.6125\n",
      "Current adversarial train loss: 1.1793110370635986\n",
      "\n",
      "Total adversarial train accuarcy: 53.35\n",
      "Total adversarial train loss: 468.34811371564865\n",
      "\n",
      "[ Test epoch: 47 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.82\n",
      "Current benign test loss: 0.7431883215904236\n",
      "Current adversarial test accuracy: 0.53\n",
      "Current adversarial test loss: 1.3811663389205933\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.76\n",
      "Current benign test loss: 0.7037577033042908\n",
      "Current adversarial test accuracy: 0.6\n",
      "Current adversarial test loss: 1.33487868309021\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.68\n",
      "Current benign test loss: 0.860159158706665\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.4745919704437256\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.7\n",
      "Current benign test loss: 0.7870815992355347\n",
      "Current adversarial test accuracy: 0.5\n",
      "Current adversarial test loss: 1.4087283611297607\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.76\n",
      "Current benign test loss: 0.7635990381240845\n",
      "Current adversarial test accuracy: 0.52\n",
      "Current adversarial test loss: 1.4026201963424683\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.84\n",
      "Current benign test loss: 0.7410727739334106\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.4168896675109863\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.8306974768638611\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.489997386932373\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.81\n",
      "Current benign test loss: 0.676697850227356\n",
      "Current adversarial test accuracy: 0.52\n",
      "Current adversarial test loss: 1.3226590156555176\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.78\n",
      "Current benign test loss: 0.714705228805542\n",
      "Current adversarial test accuracy: 0.45\n",
      "Current adversarial test loss: 1.3947046995162964\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.79\n",
      "Current benign test loss: 0.7225196957588196\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.3783769607543945\n",
      "\n",
      "Total benign test accuarcy: 76.73\n",
      "Total adversarial test Accuarcy: 48.6\n",
      "Total benign test loss: 75.0139222741127\n",
      "Total adversarial test loss: 139.08207297325134\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 48 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.1076585054397583\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.1239025592803955\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.12045419216156\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2128496170043945\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.1680235862731934\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.0849672555923462\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.609375\n",
      "Current adversarial train loss: 1.0886765718460083\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.1655522584915161\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2830551862716675\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.2142436504364014\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.1708924770355225\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.6484375\n",
      "Current adversarial train loss: 1.006777048110962\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.1999646425247192\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2282007932662964\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.1815017461776733\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.1212224960327148\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.2534512281417847\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.1289350986480713\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.5859375\n",
      "Current adversarial train loss: 1.1773637533187866\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.21951425075531\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.2067694664001465\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.1470564603805542\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.268920660018921\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.2622469663619995\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.306280255317688\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.3494682312011719\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.6015625\n",
      "Current adversarial train loss: 1.116862416267395\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.59375\n",
      "Current adversarial train loss: 1.0725030899047852\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.609375\n",
      "Current adversarial train loss: 1.0752766132354736\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.421875\n",
      "Current adversarial train loss: 1.4261093139648438\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.240496277809143\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.237045407295227\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.1743390560150146\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.3417387008666992\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.1872016191482544\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.1117452383041382\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.1117045879364014\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2563291788101196\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.3984375\n",
      "Current adversarial train loss: 1.4509567022323608\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.525\n",
      "Current adversarial train loss: 1.2042522430419922\n",
      "\n",
      "Total adversarial train accuarcy: 53.31\n",
      "Total adversarial train loss: 466.18879395723343\n",
      "\n",
      "[ Test epoch: 48 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.7933902740478516\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.4175982475280762\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.8\n",
      "Current benign test loss: 0.7113410234451294\n",
      "Current adversarial test accuracy: 0.51\n",
      "Current adversarial test loss: 1.3493778705596924\n",
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.73\n",
      "Current benign test loss: 0.7839666604995728\n",
      "Current adversarial test accuracy: 0.39\n",
      "Current adversarial test loss: 1.3264951705932617\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.818237841129303\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.421891212463379\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.7901147603988647\n",
      "Current adversarial test accuracy: 0.51\n",
      "Current adversarial test loss: 1.426853060722351\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.786228358745575\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.485790729522705\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.69\n",
      "Current benign test loss: 0.8425402641296387\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.4504226446151733\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.6958980560302734\n",
      "Current adversarial test accuracy: 0.53\n",
      "Current adversarial test loss: 1.326443076133728\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.81\n",
      "Current benign test loss: 0.7364193201065063\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.3741008043289185\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.78\n",
      "Current benign test loss: 0.7118713855743408\n",
      "Current adversarial test accuracy: 0.48\n",
      "Current adversarial test loss: 1.3189631700515747\n",
      "\n",
      "Total benign test accuarcy: 76.37\n",
      "Total adversarial test Accuarcy: 48.14\n",
      "Total benign test loss: 75.77864772081375\n",
      "Total adversarial test loss: 137.51575124263763\n",
      "Model Saved!\n",
      "\n",
      "[ Train epoch: 49 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.1320701837539673\n",
      "\n",
      "Current batch: 10\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.201620101928711\n",
      "\n",
      "Current batch: 20\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2116150856018066\n",
      "\n",
      "Current batch: 30\n",
      "Current adversarial train accuracy: 0.5859375\n",
      "Current adversarial train loss: 1.1796752214431763\n",
      "\n",
      "Current batch: 40\n",
      "Current adversarial train accuracy: 0.5078125\n",
      "Current adversarial train loss: 1.251028299331665\n",
      "\n",
      "Current batch: 50\n",
      "Current adversarial train accuracy: 0.609375\n",
      "Current adversarial train loss: 1.1233668327331543\n",
      "\n",
      "Current batch: 60\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.2176165580749512\n",
      "\n",
      "Current batch: 70\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.146782398223877\n",
      "\n",
      "Current batch: 80\n",
      "Current adversarial train accuracy: 0.4453125\n",
      "Current adversarial train loss: 1.5251188278198242\n",
      "\n",
      "Current batch: 90\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.0756901502609253\n",
      "\n",
      "Current batch: 100\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.1946468353271484\n",
      "\n",
      "Current batch: 110\n",
      "Current adversarial train accuracy: 0.46875\n",
      "Current adversarial train loss: 1.2368073463439941\n",
      "\n",
      "Current batch: 120\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.2897107601165771\n",
      "\n",
      "Current batch: 130\n",
      "Current adversarial train accuracy: 0.5703125\n",
      "Current adversarial train loss: 1.0690045356750488\n",
      "\n",
      "Current batch: 140\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.1723071336746216\n",
      "\n",
      "Current batch: 150\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2210044860839844\n",
      "\n",
      "Current batch: 160\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.2037370204925537\n",
      "\n",
      "Current batch: 170\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.159824252128601\n",
      "\n",
      "Current batch: 180\n",
      "Current adversarial train accuracy: 0.5546875\n",
      "Current adversarial train loss: 1.1858786344528198\n",
      "\n",
      "Current batch: 190\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.1766037940979004\n",
      "\n",
      "Current batch: 200\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2134639024734497\n",
      "\n",
      "Current batch: 210\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.1561641693115234\n",
      "\n",
      "Current batch: 220\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2711901664733887\n",
      "\n",
      "Current batch: 230\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.103737235069275\n",
      "\n",
      "Current batch: 240\n",
      "Current adversarial train accuracy: 0.5390625\n",
      "Current adversarial train loss: 1.129859209060669\n",
      "\n",
      "Current batch: 250\n",
      "Current adversarial train accuracy: 0.5625\n",
      "Current adversarial train loss: 1.1051770448684692\n",
      "\n",
      "Current batch: 260\n",
      "Current adversarial train accuracy: 0.515625\n",
      "Current adversarial train loss: 1.2464429140090942\n",
      "\n",
      "Current batch: 270\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.1930551528930664\n",
      "\n",
      "Current batch: 280\n",
      "Current adversarial train accuracy: 0.484375\n",
      "Current adversarial train loss: 1.2449232339859009\n",
      "\n",
      "Current batch: 290\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.3342093229293823\n",
      "\n",
      "Current batch: 300\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.1848666667938232\n",
      "\n",
      "Current batch: 310\n",
      "Current adversarial train accuracy: 0.53125\n",
      "Current adversarial train loss: 1.2161177396774292\n",
      "\n",
      "Current batch: 320\n",
      "Current adversarial train accuracy: 0.5859375\n",
      "Current adversarial train loss: 1.1233296394348145\n",
      "\n",
      "Current batch: 330\n",
      "Current adversarial train accuracy: 0.4921875\n",
      "Current adversarial train loss: 1.2174766063690186\n",
      "\n",
      "Current batch: 340\n",
      "Current adversarial train accuracy: 0.546875\n",
      "Current adversarial train loss: 1.14431631565094\n",
      "\n",
      "Current batch: 350\n",
      "Current adversarial train accuracy: 0.4609375\n",
      "Current adversarial train loss: 1.3146411180496216\n",
      "\n",
      "Current batch: 360\n",
      "Current adversarial train accuracy: 0.5234375\n",
      "Current adversarial train loss: 1.2414220571517944\n",
      "\n",
      "Current batch: 370\n",
      "Current adversarial train accuracy: 0.578125\n",
      "Current adversarial train loss: 1.115464448928833\n",
      "\n",
      "Current batch: 380\n",
      "Current adversarial train accuracy: 0.5\n",
      "Current adversarial train loss: 1.2585067749023438\n",
      "\n",
      "Current batch: 390\n",
      "Current adversarial train accuracy: 0.5375\n",
      "Current adversarial train loss: 1.0922067165374756\n",
      "\n",
      "Total adversarial train accuarcy: 53.736\n",
      "Total adversarial train loss: 462.68988502025604\n",
      "\n",
      "[ Test epoch: 49 ]\n",
      "\n",
      "Current batch: 0\n",
      "Current benign test accuracy: 0.78\n",
      "Current benign test loss: 0.7873042225837708\n",
      "Current adversarial test accuracy: 0.5\n",
      "Current adversarial test loss: 1.4408416748046875\n",
      "\n",
      "Current batch: 10\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.7305876016616821\n",
      "Current adversarial test accuracy: 0.52\n",
      "Current adversarial test loss: 1.4489599466323853\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Current batch: 20\n",
      "Current benign test accuracy: 0.72\n",
      "Current benign test loss: 0.7688397169113159\n",
      "Current adversarial test accuracy: 0.41\n",
      "Current adversarial test loss: 1.3153892755508423\n",
      "\n",
      "Current batch: 30\n",
      "Current benign test accuracy: 0.74\n",
      "Current benign test loss: 0.7821694016456604\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.39421808719635\n",
      "\n",
      "Current batch: 40\n",
      "Current benign test accuracy: 0.76\n",
      "Current benign test loss: 0.7421520352363586\n",
      "Current adversarial test accuracy: 0.46\n",
      "Current adversarial test loss: 1.4173245429992676\n",
      "\n",
      "Current batch: 50\n",
      "Current benign test accuracy: 0.83\n",
      "Current benign test loss: 0.776124894618988\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.5018190145492554\n",
      "\n",
      "Current batch: 60\n",
      "Current benign test accuracy: 0.71\n",
      "Current benign test loss: 0.8568723201751709\n",
      "Current adversarial test accuracy: 0.42\n",
      "Current adversarial test loss: 1.5095820426940918\n",
      "\n",
      "Current batch: 70\n",
      "Current benign test accuracy: 0.77\n",
      "Current benign test loss: 0.6736545562744141\n",
      "Current adversarial test accuracy: 0.47\n",
      "Current adversarial test loss: 1.3194525241851807\n",
      "\n",
      "Current batch: 80\n",
      "Current benign test accuracy: 0.8\n",
      "Current benign test loss: 0.7575286030769348\n",
      "Current adversarial test accuracy: 0.44\n",
      "Current adversarial test loss: 1.4529756307601929\n",
      "\n",
      "Current batch: 90\n",
      "Current benign test accuracy: 0.78\n",
      "Current benign test loss: 0.7243950366973877\n",
      "Current adversarial test accuracy: 0.43\n",
      "Current adversarial test loss: 1.3409357070922852\n",
      "\n",
      "Total benign test accuarcy: 76.86\n",
      "Total adversarial test Accuarcy: 48.43\n",
      "Total benign test loss: 73.87426990270615\n",
      "Total adversarial test loss: 137.28444242477417\n",
      "Model Saved!\n"
     ]
    }
   ],
   "source": [
    "t1 = time.time()\n",
    "for epoch in range(0, 50):\n",
    "    adjust_learning_rate(optimizer, epoch)\n",
    "    train(epoch)\n",
    "    test(epoch)\n",
    "t2 = time.time()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "e8bf017d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total running time is 4:24:19.248712 \n"
     ]
    }
   ],
   "source": [
    "#RTX3070\n",
    "import datetime\n",
    "deltatime = str(datetime.timedelta(seconds=t2-t1))\n",
    "print('Total running time is %s '% deltatime)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ea47bd6a",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}