Model_03-SGD-k=3.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Model_03-SGD-K=3\n",
    "\n",
    "## Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#Mandatory imports\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import log_loss\n",
    "from sklearn.model_selection import StratifiedKFold, StratifiedShuffleSplit\n",
    "from os.path import join as opj\n",
    "from matplotlib import pyplot as plt\n",
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "import pylab\n",
    "plt.rcParams['figure.figsize'] = 10, 10\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load Input Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "train = pd.read_json(\"input/train.json\")\n",
    "target_train=train['is_iceberg']\n",
    "test = pd.read_json(\"input/test.json\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "target_train=train['is_iceberg']\n",
    "test['inc_angle']=pd.to_numeric(test['inc_angle'], errors='coerce')\n",
    "train['inc_angle']=pd.to_numeric(train['inc_angle'], errors='coerce')#We have only 133 NAs.\n",
    "train['inc_angle']=train['inc_angle'].fillna(method='pad')\n",
    "X_angle=train['inc_angle']\n",
    "test['inc_angle']=pd.to_numeric(test['inc_angle'], errors='coerce')\n",
    "X_test_angle=test['inc_angle']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Generate the training data\n",
    "Use band_1 as the first color dimension and band_2 as the second color dimension. I formatted the third color dimension using the average of band_1 and band_2."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#Generate the training data\n",
    "X_band_1=np.array([np.array(band).astype(np.float32).reshape(75, 75) for band in train[\"band_1\"]])\n",
    "X_band_2=np.array([np.array(band).astype(np.float32).reshape(75, 75) for band in train[\"band_2\"]])\n",
    "X_band_3=(X_band_1+X_band_2)/2\n",
    "#X_band_3=np.array([np.full((75, 75), angel).astype(np.float32) for angel in train[\"inc_angle\"]])\n",
    "X_train = np.concatenate([X_band_1[:, :, :, np.newaxis]\n",
    "                          , X_band_2[:, :, :, np.newaxis]\n",
    "                         , X_band_3[:, :, :, np.newaxis]], axis=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X_band_test_1=np.array([np.array(band).astype(np.float32).reshape(75, 75) for band in test[\"band_1\"]])\n",
    "X_band_test_2=np.array([np.array(band).astype(np.float32).reshape(75, 75) for band in test[\"band_2\"]])\n",
    "X_band_test_3=(X_band_test_1+X_band_test_2)/2\n",
    "#X_band_test_3=np.array([np.full((75, 75), angel).astype(np.float32) for angel in test[\"inc_angle\"]])\n",
    "X_test = np.concatenate([X_band_test_1[:, :, :, np.newaxis]\n",
    "                          , X_band_test_2[:, :, :, np.newaxis]\n",
    "                         , X_band_test_3[:, :, :, np.newaxis]], axis=-1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Transfer Learning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#Import Keras.\n",
    "from matplotlib import pyplot\n",
    "from keras.optimizers import RMSprop\n",
    "# from keras.preprocessing.image import ImageDataGenerator\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Conv2D, MaxPooling2D, Dense, Dropout, Input, Flatten, Activation\n",
    "from keras.layers import GlobalMaxPooling2D\n",
    "from keras.layers.normalization import BatchNormalization\n",
    "from keras.layers.merge import Concatenate\n",
    "from keras.models import Model\n",
    "from keras import initializers\n",
    "from keras.optimizers import Adam\n",
    "from keras.optimizers import rmsprop\n",
    "from keras.optimizers import SGD\n",
    "from keras.optimizers import Nadam\n",
    "from keras.optimizers import Adadelta\n",
    "from keras.layers.advanced_activations import LeakyReLU, PReLU\n",
    "from keras.callbacks import ModelCheckpoint, Callback, EarlyStopping\n",
    "\n",
    "from keras.datasets import cifar10\n",
    "from keras.applications.inception_v3 import InceptionV3\n",
    "from keras.applications.vgg16 import VGG16\n",
    "from keras.applications.xception import Xception\n",
    "from keras.applications.mobilenet import MobileNet\n",
    "from keras.applications.vgg19 import VGG19\n",
    "from keras.layers import Concatenate, Dense, LSTM, Input, concatenate\n",
    "from keras.preprocessing import image\n",
    "from keras.applications.vgg16 import preprocess_input"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Image Transformation\n",
    "Use the code below to perform the data augmentation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#Data Aug for multi-input\n",
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "batch_size=64\n",
    "# Define the image transformations here\n",
    "gen = ImageDataGenerator(horizontal_flip = True,\n",
    "                         vertical_flip = True,\n",
    "                         width_shift_range = 0.,\n",
    "                         height_shift_range = 0.,\n",
    "                         channel_shift_range=0,\n",
    "                         zoom_range = 0.2,\n",
    "                         rotation_range = 10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Merge two generators"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Here is the function that merges our two generators\n",
    "# We use the exact same generator with the same random seed for both the y and angle arrays\n",
    "def gen_flow_for_two_inputs(X1, X2, y):\n",
    "    genX1 = gen.flow(X1,y,  batch_size=batch_size,seed=55)\n",
    "    genX2 = gen.flow(X1,X2, batch_size=batch_size,seed=55)\n",
    "    while True:\n",
    "            X1i = genX1.next()\n",
    "            X2i = genX2.next()\n",
    "            #Assert arrays are equal - this was for peace of mind, but slows down training\n",
    "            #np.testing.assert_array_equal(X1i[0],X2i[0])\n",
    "            yield [X1i[0], X2i[1]], X1i[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Finally create generator\n",
    "def get_callbacks(filepath, patience=2):\n",
    "    es = EarlyStopping('val_loss', patience=10, mode=\"min\")\n",
    "    msave = ModelCheckpoint(filepath, save_best_only=True)\n",
    "    return [es, msave]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## VGG16 Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def getVggAngleModel():\n",
    "    input_2 = Input(shape=[1], name=\"angle\")\n",
    "    angle_layer = Dense(1, )(input_2)\n",
    "    base_model = VGG16(weights='imagenet', include_top=False, \n",
    "                 input_shape=X_train.shape[1:], classes=1)\n",
    "    x = base_model.get_layer('block5_pool').output\n",
    "    \n",
    "\n",
    "    x = GlobalMaxPooling2D()(x)\n",
    "    merge_one = concatenate([x, angle_layer])\n",
    "    merge_one = Dense(512, activation='relu', name='fc2')(merge_one)\n",
    "    merge_one = Dropout(0.3)(merge_one)\n",
    "    merge_one = Dense(512, activation='relu', name='fc3')(merge_one)\n",
    "    merge_one = Dropout(0.3)(merge_one)\n",
    "    \n",
    "    predictions = Dense(1, activation='sigmoid')(merge_one)\n",
    "    \n",
    "    model = Model(inputs=[base_model.input, input_2], outputs=predictions)\n",
    "    \n",
    "#     adadelta= Adadelta(lr=1.0, rho=0.95, epsilon=None, decay=0.0)\n",
    "#     model.compile(loss='binary_crossentropy',\n",
    "#                   optimizer=adadelta,\n",
    "#                   metrics=['accuracy'])\n",
    "    \n",
    "    sgd = SGD(lr=1e-3, decay=1e-6, momentum=0.9, nesterov=True)\n",
    "    model.compile(loss='binary_crossentropy',\n",
    "                  optimizer=sgd,\n",
    "                  metrics=['accuracy'])\n",
    "    return model\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Stratified K-Fold Cross Validation\n",
    "### K=3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "===================FOLD= 0\n",
      "Epoch 1/50\n",
      "24/24 [==============================] - 15s 606ms/step - loss: 0.6458 - acc: 0.6254 - val_loss: 0.4875 - val_acc: 0.7364\n",
      "Epoch 2/50\n",
      "24/24 [==============================] - 13s 541ms/step - loss: 0.4244 - acc: 0.7987 - val_loss: 0.2713 - val_acc: 0.8579\n",
      "Epoch 3/50\n",
      "24/24 [==============================] - 13s 536ms/step - loss: 0.3548 - acc: 0.8366 - val_loss: 0.2634 - val_acc: 0.8804\n",
      "Epoch 4/50\n",
      "24/24 [==============================] - 13s 542ms/step - loss: 0.2957 - acc: 0.8691 - val_loss: 0.2435 - val_acc: 0.8879\n",
      "Epoch 5/50\n",
      "24/24 [==============================] - 12s 519ms/step - loss: 0.2796 - acc: 0.8734 - val_loss: 0.2443 - val_acc: 0.8841\n",
      "Epoch 6/50\n",
      "24/24 [==============================] - 13s 543ms/step - loss: 0.2788 - acc: 0.8745 - val_loss: 0.2226 - val_acc: 0.8953\n",
      "Epoch 7/50\n",
      "24/24 [==============================] - 13s 543ms/step - loss: 0.2378 - acc: 0.8995 - val_loss: 0.2053 - val_acc: 0.9121\n",
      "Epoch 8/50\n",
      "24/24 [==============================] - 13s 538ms/step - loss: 0.2570 - acc: 0.8880 - val_loss: 0.1965 - val_acc: 0.9178\n",
      "Epoch 9/50\n",
      "24/24 [==============================] - 13s 544ms/step - loss: 0.2157 - acc: 0.9093 - val_loss: 0.1937 - val_acc: 0.9121\n",
      "Epoch 10/50\n",
      "24/24 [==============================] - 12s 520ms/step - loss: 0.2081 - acc: 0.9172 - val_loss: 0.2074 - val_acc: 0.9121\n",
      "Epoch 11/50\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1842 - acc: 0.9269 - val_loss: 0.2232 - val_acc: 0.9084\n",
      "Epoch 12/50\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2175 - acc: 0.9061 - val_loss: 0.1978 - val_acc: 0.9252\n",
      "Epoch 13/50\n",
      "24/24 [==============================] - 13s 521ms/step - loss: 0.2026 - acc: 0.9099 - val_loss: 0.2022 - val_acc: 0.9234\n",
      "Epoch 14/50\n",
      "24/24 [==============================] - 13s 545ms/step - loss: 0.2000 - acc: 0.9203 - val_loss: 0.1870 - val_acc: 0.9271\n",
      "Epoch 15/50\n",
      "24/24 [==============================] - 13s 521ms/step - loss: 0.1908 - acc: 0.9232 - val_loss: 0.2087 - val_acc: 0.9121\n",
      "Epoch 16/50\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1899 - acc: 0.9200 - val_loss: 0.2005 - val_acc: 0.9215\n",
      "Epoch 17/50\n",
      "24/24 [==============================] - 14s 574ms/step - loss: 0.1742 - acc: 0.9287 - val_loss: 0.1788 - val_acc: 0.9271\n",
      "Epoch 18/50\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1818 - acc: 0.9196 - val_loss: 0.2432 - val_acc: 0.9140\n",
      "Epoch 19/50\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1567 - acc: 0.9444 - val_loss: 0.2032 - val_acc: 0.9234\n",
      "Epoch 20/50\n",
      "24/24 [==============================] - 13s 523ms/step - loss: 0.1714 - acc: 0.9372 - val_loss: 0.1972 - val_acc: 0.9234\n",
      "Epoch 21/50\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1444 - acc: 0.9377 - val_loss: 0.2019 - val_acc: 0.9252\n",
      "Epoch 22/50\n",
      "24/24 [==============================] - 13s 523ms/step - loss: 0.1556 - acc: 0.9404 - val_loss: 0.2078 - val_acc: 0.9252\n",
      "Epoch 23/50\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1488 - acc: 0.9399 - val_loss: 0.2274 - val_acc: 0.9084\n",
      "Epoch 24/50\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1417 - acc: 0.9380 - val_loss: 0.2315 - val_acc: 0.9140\n",
      "Epoch 25/50\n",
      "24/24 [==============================] - 13s 524ms/step - loss: 0.1621 - acc: 0.9328 - val_loss: 0.2078 - val_acc: 0.9327\n",
      "Epoch 26/50\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1266 - acc: 0.9523 - val_loss: 0.2469 - val_acc: 0.9084\n",
      "Epoch 27/50\n",
      "24/24 [==============================] - 13s 522ms/step - loss: 0.1206 - acc: 0.9492 - val_loss: 0.2615 - val_acc: 0.9084\n",
      "Train loss: 0.126367402938\n",
      "Train accuracy: 0.948550047553\n",
      "Test loss: 0.17882133778\n",
      "Test accuracy: 0.927102805521\n",
      "\n",
      "===================FOLD= 1\n",
      "Epoch 1/50\n",
      "24/24 [==============================] - 15s 620ms/step - loss: 0.6525 - acc: 0.6571 - val_loss: 0.9360 - val_acc: 0.5832\n",
      "Epoch 2/50\n",
      "24/24 [==============================] - 13s 548ms/step - loss: 0.4921 - acc: 0.7454 - val_loss: 0.3282 - val_acc: 0.8598\n",
      "Epoch 3/50\n",
      "24/24 [==============================] - 13s 542ms/step - loss: 0.3726 - acc: 0.8147 - val_loss: 0.2801 - val_acc: 0.8710\n",
      "Epoch 4/50\n",
      "24/24 [==============================] - 13s 545ms/step - loss: 0.3160 - acc: 0.8422 - val_loss: 0.2626 - val_acc: 0.8804\n",
      "Epoch 5/50\n",
      "24/24 [==============================] - 13s 541ms/step - loss: 0.2870 - acc: 0.8626 - val_loss: 0.2292 - val_acc: 0.9159\n",
      "Epoch 6/50\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2413 - acc: 0.8896 - val_loss: 0.2397 - val_acc: 0.9009\n",
      "Epoch 7/50\n",
      "24/24 [==============================] - 13s 545ms/step - loss: 0.2762 - acc: 0.8637 - val_loss: 0.2129 - val_acc: 0.9178\n",
      "Epoch 8/50\n",
      "24/24 [==============================] - 12s 521ms/step - loss: 0.2376 - acc: 0.8943 - val_loss: 0.2233 - val_acc: 0.9215\n",
      "Epoch 9/50\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2293 - acc: 0.8915 - val_loss: 0.2162 - val_acc: 0.9140\n",
      "Epoch 10/50\n",
      "24/24 [==============================] - 13s 521ms/step - loss: 0.2270 - acc: 0.8982 - val_loss: 0.2391 - val_acc: 0.9103\n",
      "Epoch 11/50\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.2393 - acc: 0.8928 - val_loss: 0.2349 - val_acc: 0.9234\n",
      "Epoch 12/50\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1953 - acc: 0.9117 - val_loss: 0.2132 - val_acc: 0.9215\n",
      "Epoch 13/50\n",
      "24/24 [==============================] - 13s 523ms/step - loss: 0.1784 - acc: 0.9243 - val_loss: 0.2833 - val_acc: 0.8953\n",
      "Epoch 14/50\n",
      "24/24 [==============================] - 13s 531ms/step - loss: 0.2171 - acc: 0.9050 - val_loss: 0.2139 - val_acc: 0.9159\n",
      "Epoch 15/50\n",
      "24/24 [==============================] - 13s 545ms/step - loss: 0.2166 - acc: 0.9047 - val_loss: 0.1977 - val_acc: 0.9140\n",
      "Epoch 16/50\n",
      "24/24 [==============================] - 13s 530ms/step - loss: 0.1830 - acc: 0.9259 - val_loss: 0.2255 - val_acc: 0.9121\n",
      "Epoch 17/50\n",
      "24/24 [==============================] - 13s 549ms/step - loss: 0.1904 - acc: 0.9134 - val_loss: 0.1915 - val_acc: 0.9346\n",
      "Epoch 18/50\n",
      "24/24 [==============================] - 13s 528ms/step - loss: 0.1696 - acc: 0.9338 - val_loss: 0.1995 - val_acc: 0.9234\n",
      "Epoch 19/50\n",
      "24/24 [==============================] - 13s 528ms/step - loss: 0.1875 - acc: 0.9174 - val_loss: 0.1945 - val_acc: 0.9234\n",
      "Epoch 20/50\n",
      "24/24 [==============================] - 13s 523ms/step - loss: 0.1612 - acc: 0.9296 - val_loss: 0.2056 - val_acc: 0.9271\n",
      "Epoch 21/50\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1804 - acc: 0.9236 - val_loss: 0.2664 - val_acc: 0.9103\n",
      "Epoch 22/50\n",
      "24/24 [==============================] - 13s 521ms/step - loss: 0.1854 - acc: 0.9219 - val_loss: 0.2163 - val_acc: 0.9140\n",
      "Epoch 23/50\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1555 - acc: 0.9382 - val_loss: 0.2222 - val_acc: 0.9234\n",
      "Epoch 24/50\n",
      "24/24 [==============================] - 13s 526ms/step - loss: 0.1448 - acc: 0.9390 - val_loss: 0.2441 - val_acc: 0.9084\n",
      "Epoch 25/50\n",
      "24/24 [==============================] - 13s 522ms/step - loss: 0.1477 - acc: 0.9407 - val_loss: 0.1950 - val_acc: 0.9252\n",
      "Epoch 26/50\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1474 - acc: 0.9371 - val_loss: 0.2337 - val_acc: 0.9178\n",
      "Epoch 27/50\n",
      "24/24 [==============================] - 13s 523ms/step - loss: 0.1481 - acc: 0.9419 - val_loss: 0.2342 - val_acc: 0.9084\n",
      "Train loss: 0.13941859417\n",
      "Train accuracy: 0.941066417993\n",
      "Test loss: 0.191542158394\n",
      "Test accuracy: 0.934579440589\n",
      "\n",
      "===================FOLD= 2\n",
      "Epoch 1/50\n",
      "24/24 [==============================] - 17s 715ms/step - loss: 0.6872 - acc: 0.6379 - val_loss: 0.4193 - val_acc: 0.7921\n",
      "Epoch 2/50\n",
      "24/24 [==============================] - 13s 547ms/step - loss: 0.4893 - acc: 0.7780 - val_loss: 0.3463 - val_acc: 0.8202\n",
      "Epoch 3/50\n",
      "24/24 [==============================] - 13s 541ms/step - loss: 0.3445 - acc: 0.8450 - val_loss: 0.3269 - val_acc: 0.8633\n",
      "Epoch 4/50\n",
      "24/24 [==============================] - 13s 548ms/step - loss: 0.3075 - acc: 0.8727 - val_loss: 0.3096 - val_acc: 0.8652\n",
      "Epoch 5/50\n",
      "24/24 [==============================] - 13s 546ms/step - loss: 0.2691 - acc: 0.8800 - val_loss: 0.2888 - val_acc: 0.8670\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 6/50\n",
      "24/24 [==============================] - 13s 551ms/step - loss: 0.2541 - acc: 0.8922 - val_loss: 0.2709 - val_acc: 0.8914\n",
      "Epoch 7/50\n",
      "24/24 [==============================] - 13s 550ms/step - loss: 0.2282 - acc: 0.9030 - val_loss: 0.2663 - val_acc: 0.8933\n",
      "Epoch 8/50\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.2270 - acc: 0.9026 - val_loss: 0.2804 - val_acc: 0.8839\n",
      "Epoch 9/50\n",
      "24/24 [==============================] - 13s 531ms/step - loss: 0.2131 - acc: 0.9137 - val_loss: 0.2776 - val_acc: 0.8801\n",
      "Epoch 10/50\n",
      "24/24 [==============================] - 13s 548ms/step - loss: 0.2104 - acc: 0.9150 - val_loss: 0.2506 - val_acc: 0.9026\n",
      "Epoch 11/50\n",
      "24/24 [==============================] - 13s 531ms/step - loss: 0.1820 - acc: 0.9252 - val_loss: 0.3584 - val_acc: 0.8558\n",
      "Epoch 12/50\n",
      "24/24 [==============================] - 13s 530ms/step - loss: 0.2070 - acc: 0.9185 - val_loss: 0.2569 - val_acc: 0.8914\n",
      "Epoch 13/50\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1892 - acc: 0.9226 - val_loss: 0.2548 - val_acc: 0.8876\n",
      "Epoch 14/50\n",
      "24/24 [==============================] - 13s 529ms/step - loss: 0.1682 - acc: 0.9366 - val_loss: 0.2647 - val_acc: 0.8895\n",
      "Epoch 15/50\n",
      "24/24 [==============================] - 13s 525ms/step - loss: 0.1703 - acc: 0.9290 - val_loss: 0.2749 - val_acc: 0.8801\n",
      "Epoch 16/50\n",
      "24/24 [==============================] - 13s 528ms/step - loss: 0.1846 - acc: 0.9294 - val_loss: 0.2508 - val_acc: 0.9064\n",
      "Epoch 17/50\n",
      "24/24 [==============================] - 13s 550ms/step - loss: 0.1732 - acc: 0.9249 - val_loss: 0.2878 - val_acc: 0.8633\n",
      "Epoch 18/50\n",
      "24/24 [==============================] - 13s 527ms/step - loss: 0.1647 - acc: 0.9426 - val_loss: 0.2565 - val_acc: 0.8933\n",
      "Epoch 19/50\n",
      "24/24 [==============================] - 13s 528ms/step - loss: 0.1660 - acc: 0.9410 - val_loss: 0.2550 - val_acc: 0.8839\n",
      "Epoch 20/50\n",
      "24/24 [==============================] - 13s 523ms/step - loss: 0.1537 - acc: 0.9381 - val_loss: 0.2626 - val_acc: 0.8858\n",
      "Train loss: 0.156089660163\n",
      "Train accuracy: 0.943925234091\n",
      "Test loss: 0.250586017799\n",
      "Test accuracy: 0.902621723293\n",
      "\n",
      " Train Log Loss Validation=  0.147466692344\n",
      " Test Log Loss Validation=  0.206955990373\n"
     ]
    }
   ],
   "source": [
    "#def myAngleCV(X_train, X_angle, X_test):\n",
    "K=3\n",
    "folds = list(StratifiedKFold(n_splits=K, shuffle=True, random_state=16).split(X_train, target_train))\n",
    "y_test_pred_log = 0\n",
    "y_train_pred_log=0\n",
    "y_valid_pred_log = 0.0*target_train\n",
    "for j, (train_idx, test_idx) in enumerate(folds):\n",
    "    print('\\n===================FOLD=',j)\n",
    "    X_train_cv = X_train[train_idx]\n",
    "    y_train_cv = target_train[train_idx]\n",
    "    X_holdout = X_train[test_idx]\n",
    "    Y_holdout= target_train[test_idx]\n",
    "    \n",
    "    #Angle\n",
    "    X_angle_cv=X_angle[train_idx]\n",
    "    X_angle_hold=X_angle[test_idx]\n",
    "    #define file path and get callbacks\n",
    "    file_path = \"%s_aug_model_weights.hdf5\"%j\n",
    "    callbacks = get_callbacks(filepath=file_path, patience=5)\n",
    "    gen_flow = gen_flow_for_two_inputs(X_train_cv, X_angle_cv, y_train_cv)\n",
    "    galaxyModel= getVggAngleModel()\n",
    "    histr = galaxyModel.fit_generator(\n",
    "            gen_flow,\n",
    "            steps_per_epoch=24,\n",
    "            epochs=50,\n",
    "            shuffle=True,\n",
    "            verbose=1,\n",
    "            validation_data=([X_holdout,X_angle_hold], Y_holdout),\n",
    "            callbacks=callbacks)\n",
    "     #Getting the Best Model\n",
    "    galaxyModel.load_weights(filepath=file_path)\n",
    "    #Getting Training Score\n",
    "    score = galaxyModel.evaluate([X_train_cv,X_angle_cv], y_train_cv, verbose=0)\n",
    "    print('Train loss:', score[0])\n",
    "    print('Train accuracy:', score[1])\n",
    "    #Getting Test Score\n",
    "    score = galaxyModel.evaluate([X_holdout,X_angle_hold], Y_holdout, verbose=0)\n",
    "    print('Test loss:', score[0])\n",
    "    print('Test accuracy:', score[1])\n",
    "\n",
    "    #Getting validation Score.\n",
    "    pred_valid=galaxyModel.predict([X_holdout,X_angle_hold])\n",
    "    y_valid_pred_log[test_idx] = pred_valid.reshape(pred_valid.shape[0])\n",
    "\n",
    "    #Getting Test Scores\n",
    "    temp_test=galaxyModel.predict([X_test, X_test_angle])\n",
    "    y_test_pred_log+=temp_test.reshape(temp_test.shape[0])\n",
    "\n",
    "    #Getting Train Scores\n",
    "    temp_train=galaxyModel.predict([X_train, X_angle])\n",
    "    y_train_pred_log+=temp_train.reshape(temp_train.shape[0])\n",
    "\n",
    "y_test_pred_log=y_test_pred_log/K\n",
    "y_train_pred_log=y_train_pred_log/K\n",
    "\n",
    "print('\\n Train Log Loss Validation= ',log_loss(target_train, y_train_pred_log))\n",
    "print(' Test Log Loss Validation= ',log_loss(target_train, y_valid_pred_log))\n",
    "#return y_test_pred_log"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#preds=myAngleCV(X_train, X_angle, X_test)\n",
    "preds = y_test_pred_log"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Submission to kaggle\n",
    "submission = pd.DataFrame()\n",
    "submission['id']=test['id']\n",
    "submission['is_iceberg']=preds\n",
    "submission.to_csv('sub_SGD.csv', index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model Accuracy & Model Loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_keys(['loss', 'val_acc', 'val_loss', 'acc'])\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAEWCAYAAACXGLsWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7ff8008e69b0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEWCAYAAACJ0YulAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7ff800807dd8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# list all data in history\n",
    "print(histr.history.keys())\n",
    "# summarize history for accuracy\n",
    "plt.plot(histr.history['acc'])\n",
    "plt.plot(histr.history['val_acc'])\n",
    "plt.title('model accuracy')\n",
    "plt.ylabel('accuracy')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'test'], loc='upper left')\n",
    "plt.show()\n",
    "# summarize history for loss\n",
    "plt.plot(histr.history['loss'])\n",
    "plt.plot(histr.history['val_loss'])\n",
    "plt.title('model loss')\n",
    "plt.ylabel('loss')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'test'], loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Prediction Result\n",
    "### Compare actual and predicted values with some validation set images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import matplotlib.image as mpimg\n",
    "\n",
    "min1 = np.min(X_holdout)\n",
    "max1 = np.max(X_holdout)\n",
    "diff1 = max1 - min1\n",
    "my_sum = lambda x: (x - min1)/diff1\n",
    "\n",
    "X_holdout2 = my_sum(X_holdout)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(534, 75, 75, 3)"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_holdout2.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x7ff7ff668cf8>"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUQAAAEICAYAAAAncI3RAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7ff7ff7a2d30>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X_holdout2[0], cmap='gray', interpolation='nearest');\n",
    "tl = \"Actual label : \" +str(Y_holdout[9])+ \",\"+\" iceberg_probability : \"+str(pred_valid[0])\n",
    "plt.title(tl)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "9       0\n",
       "13      1\n",
       "14      0\n",
       "19      1\n",
       "26      1\n",
       "27      0\n",
       "28      1\n",
       "36      1\n",
       "38      1\n",
       "42      0\n",
       "43      0\n",
       "44      0\n",
       "52      0\n",
       "63      0\n",
       "73      1\n",
       "76      0\n",
       "77      1\n",
       "79      0\n",
       "80      0\n",
       "81      1\n",
       "83      1\n",
       "86      1\n",
       "89      0\n",
       "92      0\n",
       "93      0\n",
       "95      0\n",
       "96      1\n",
       "97      1\n",
       "100     0\n",
       "101     0\n",
       "       ..\n",
       "1533    0\n",
       "1537    0\n",
       "1539    0\n",
       "1541    0\n",
       "1542    0\n",
       "1543    0\n",
       "1545    0\n",
       "1548    0\n",
       "1549    0\n",
       "1552    0\n",
       "1553    0\n",
       "1559    0\n",
       "1561    0\n",
       "1566    0\n",
       "1567    0\n",
       "1569    0\n",
       "1576    0\n",
       "1577    0\n",
       "1579    0\n",
       "1585    0\n",
       "1587    0\n",
       "1588    0\n",
       "1590    0\n",
       "1591    0\n",
       "1596    0\n",
       "1597    0\n",
       "1599    0\n",
       "1601    0\n",
       "1602    0\n",
       "1603    0\n",
       "Name: is_iceberg, Length: 534, dtype: int64"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y_holdout"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x7ff7ff5b1eb8>"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAT0AAAEICAYAAAAtLCODAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nZ+w66HcAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7ff7ff58f588>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X_holdout2[2], cmap='gray', interpolation='nearest');\n",
    "tl = \"Actual label : \" +str(Y_holdout[14])+ \",\"+\" iceberg_probability : \"+str(pred_valid[2])\n",
    "plt.title(tl)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(534, 1)"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred_valid.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(534,)"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y_holdout.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x7ff7ff30d438>"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUUAAAEICAYAAADIsubvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7ff7ff39e278>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X_holdout2[7], cmap='gray', interpolation='nearest');\n",
    "tl = \"Actual label : \" +str(Y_holdout[28])+ \",\"+\" iceberg_probability : \"+str(pred_valid[7])\n",
    "plt.title(tl)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x7ff801059c50>"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUUAAAEICAYAAADIsubvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7ff7ff2f4fd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X_holdout2[532], cmap='gray', interpolation='nearest');\n",
    "tl = \"Actual label : \" +str(Y_holdout[1602])+ \",\"+\" iceberg_probability : \"+str(pred_valid[532])\n",
    "plt.title(tl)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x7ff7ff71e780>"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUUAAAEICAYAAADIsubvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7ff8008932e8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X_holdout2[15], cmap='gray', interpolation='nearest');\n",
    "tl = \"Actual label : \" +str(Y_holdout[73])+ \",\"+\" iceberg_probability : \"+str(pred_valid[15])\n",
    "plt.title(tl)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}