1_measure_files/Automatic_subj_training.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "ccc04654",
   "metadata": {},
   "source": [
    "## Var name setting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "d887cfd2",
   "metadata": {},
   "outputs": [],
   "source": [
    "subjectID='3'\n",
    "noEpochs=1000\n",
    "noLSTMUnits=100\n",
    "\n",
    "# Change to true if input file 's features are sorted by time (False by default) \n",
    "orderedInput=True\n",
    "\n",
    "# Change to 1 if want to export to CSV O2 precict against test (0 by default)\n",
    "exportCSV=0\n",
    "\n",
    "# Chabge to 1 if want to save the training model (0 by default)\n",
    "savedModel=0\n",
    "\n",
    "# Set seed value for uniform (NOT RECOMMEND TO CHANGE, by default 12321)\n",
    "seed_value = 12321\n",
    "\n",
    "# Setting name of output files \n",
    "if orderedInput:\n",
    "    fileName='inputFeatures'+subjectID+'_sorted.mat'\n",
    "    modelSaveName='Subj-'+subjectID+'-'+str(noEpochs)+'pochs-order-'+str(noLSTMUnits)+'-LSTM_u.h5'\n",
    "    figPredictName='Subj'+subjectID+'-'+str(noEpochs)+'pochs-order-'+str(noLSTMUnits)+'-LSTM_u.jpg'\n",
    "    figRMSEName='RMSE_Subj'+subjectID+'-'+str(noEpochs)+'pochs-order-'+str(noLSTMUnits)+'-LSTM_u.jpg'\n",
    "\n",
    "else:\n",
    "    \n",
    "    # Unordered input features\n",
    "    fileName='inputFeatures'+subjectID+'.mat'\n",
    "    modelSaveName='Subj-'+subjectID+'-'+str(noEpochs)+'pochs-'+str(noLSTMUnits)+'-LSTM_u.h5'\n",
    "    figPredictName='Subj'+subjectID+'-'+str(noEpochs)+'pochs-'+str(noLSTMUnits)+'-LSTM_u.jpg'\n",
    "    figRMSEName='RMSE_Subj'+subjectID+'-'+str(noEpochs)+'pochs-'+str(noLSTMUnits)+'-LSTM_u.jpg'"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d64db4a1",
   "metadata": {},
   "source": [
    "## Include lib"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "6f273d91",
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.models import load_model\n",
    "import scipy.io as sio \n",
    "import tensorflow as tf\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import os\n",
    "import random\n",
    "\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from tensorflow.keras.models import Model\n",
    "from keras.models import Sequential\n",
    "from keras.optimizers import Adam\n",
    "from keras.layers import LSTM\n",
    "from keras.layers import Dense\n",
    "from keras.layers import Input, Bidirectional\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from tensorflow.keras.models import Model\n",
    "from tensorflow.python.client import device_lib\n",
    "from datetime import datetime as dt\n",
    "from keras import backend as K\n",
    "#from comet_ml import Experiment"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d232b155",
   "metadata": {},
   "source": [
    "## Set seed for training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "d846f5ad",
   "metadata": {},
   "outputs": [],
   "source": [
    "os.environ['PYTHONHASHSEED']=str(seed_value)\n",
    "random.seed(seed_value)\n",
    "np.random.seed(seed_value)\n",
    "tf.random.set_seed(seed_value)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a9410f08",
   "metadata": {},
   "source": [
    "## Read input from Matlab file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "9be7b2e8",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "gpu_devices = tf.config.experimental.list_physical_devices('GPU')\n",
    "for device in gpu_devices:\n",
    "    tf.config.experimental.set_memory_growth(device, True)\n",
    "\n",
    "i=sio.loadmat(fileName)\n",
    "\n",
    "TTest=i['TTest']\n",
    "TTrain=i['TTrain']\n",
    "XTrain=i['XTrain']\n",
    "XTest=i['XTest']\n",
    "\n",
    "XTrain=XTrain.reshape(-1)\n",
    "TTrain=TTrain.reshape(-1)\n",
    "XTest=XTest.reshape(-1)\n",
    "TTest=TTest.reshape(-1)\n",
    "\n",
    "test=np.transpose(XTrain).reshape(-1)\n",
    "test2=np.transpose(XTest).reshape(-1)\n",
    "\n",
    "list_test=[]\n",
    "for i in range(len(test)):\n",
    "    test[i]=np.transpose(np.array(test[i]))\n",
    "    list_test.append(test[i])\n",
    "\n",
    "for j in range(len(test2)):  \n",
    "    test2[j]=np.transpose(np.array(test2[j]))\n",
    "    list_test.append(test2[j])  \n",
    "\n",
    "    \n",
    "list_test=np.vstack(list_test)\n",
    "scaler = StandardScaler()\n",
    "scaler=scaler.fit(list_test)\n",
    "scaled_list_test=scaler.transform(list_test)\n",
    "\n",
    "XT_s_stack=list()\n",
    "i=0\n",
    "while i+50 <= (len(scaled_list_test)):\n",
    "    XT_s_stack.append(scaled_list_test[i:i+50])\n",
    "    i=i+50\n",
    "\n",
    "\n",
    "XT_s_stack=np.stack(XT_s_stack)\n",
    "\n",
    "cutPoint=np.shape(TTest.reshape(-1))[0]\n",
    "\n",
    "XTrain_s=XT_s_stack[:-cutPoint,:,:]\n",
    "XTest_s=XT_s_stack[-cutPoint:,:,:]\n",
    "\n",
    "Test=np.transpose(TTest)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "637ca220",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'\\nprint(list_test)\\nHR_arr=[]\\nfor i in list_test:\\n    HR_arr.append(i[0][4])\\n    HR_arr.append(i[1][4])\\n    \\nimport matplotlib.pyplot as plt\\nprint(HR_arr)\\nplt.plot(HR_arr)\\nplt.show()\\n'"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#fileNameHR=\"5inputs.mat\"\n",
    "i=sio.loadmat(fileName)\n",
    "\n",
    "XTrain=i['XTrain']\n",
    "XTrain=XTrain.reshape(-1)\n",
    "\n",
    "#print(np.shape(XTrain))\n",
    "test=np.transpose(XTrain).reshape(-1)\n",
    "\n",
    "list_test=[]\n",
    "for i in range(len(test)):\n",
    "    test[i]=np.transpose(np.array(test[i]))\n",
    "    list_test.append(test[i])\n",
    "\n",
    "    \n",
    "'''\n",
    "print(list_test)\n",
    "HR_arr=[]\n",
    "for i in list_test:\n",
    "    HR_arr.append(i[0][4])\n",
    "    HR_arr.append(i[1][4])\n",
    "    \n",
    "import matplotlib.pyplot as plt\n",
    "print(HR_arr)\n",
    "plt.plot(HR_arr)\n",
    "plt.show()\n",
    "'''"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "55305abc",
   "metadata": {},
   "source": [
    "## Configure the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "91546b25",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " lstm (LSTM)                 (None, 100)               42000     \n",
      "                                                                 \n",
      " dense (Dense)               (None, 1)                 101       \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 42,101\n",
      "Trainable params: 42,101\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "length = 50\n",
    "n_features = 4\n",
    "model = Sequential()\n",
    "model.add(LSTM(noLSTMUnits,input_shape=(length,n_features)))\n",
    "model.add(Dense(1))\n",
    "\n",
    "model.compile(optimizer='adam', loss='mean_squared_logarithmic_error')\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "66f31e4d",
   "metadata": {},
   "source": [
    "## Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "9a62b3fc",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/1000\n",
      "40/40 - 3s - loss: 4.9695 - 3s/epoch - 78ms/step\n",
      "Epoch 2/1000\n",
      "40/40 - 0s - loss: 3.0196 - 317ms/epoch - 8ms/step\n",
      "Epoch 3/1000\n",
      "40/40 - 0s - loss: 0.7536 - 342ms/epoch - 9ms/step\n",
      "Epoch 4/1000\n",
      "40/40 - 0s - loss: 0.2792 - 329ms/epoch - 8ms/step\n",
      "Epoch 5/1000\n",
      "40/40 - 0s - loss: 0.2626 - 312ms/epoch - 8ms/step\n",
      "Epoch 6/1000\n",
      "40/40 - 0s - loss: 0.2537 - 342ms/epoch - 9ms/step\n",
      "Epoch 7/1000\n",
      "40/40 - 0s - loss: 0.2435 - 355ms/epoch - 9ms/step\n",
      "Epoch 8/1000\n",
      "40/40 - 0s - loss: 0.2393 - 315ms/epoch - 8ms/step\n",
      "Epoch 9/1000\n",
      "40/40 - 0s - loss: 0.2368 - 315ms/epoch - 8ms/step\n",
      "Epoch 10/1000\n",
      "40/40 - 0s - loss: 0.2348 - 338ms/epoch - 8ms/step\n",
      "Epoch 11/1000\n",
      "40/40 - 0s - loss: 0.2332 - 383ms/epoch - 10ms/step\n",
      "Epoch 12/1000\n",
      "40/40 - 0s - loss: 0.2316 - 362ms/epoch - 9ms/step\n",
      "Epoch 13/1000\n",
      "40/40 - 0s - loss: 0.2455 - 370ms/epoch - 9ms/step\n",
      "Epoch 14/1000\n",
      "40/40 - 0s - loss: 0.2450 - 357ms/epoch - 9ms/step\n",
      "Epoch 15/1000\n",
      "40/40 - 0s - loss: 0.2416 - 363ms/epoch - 9ms/step\n",
      "Epoch 16/1000\n",
      "40/40 - 0s - loss: 0.2379 - 291ms/epoch - 7ms/step\n",
      "Epoch 17/1000\n",
      "40/40 - 0s - loss: 0.2328 - 339ms/epoch - 8ms/step\n",
      "Epoch 18/1000\n",
      "40/40 - 0s - loss: 0.2261 - 323ms/epoch - 8ms/step\n",
      "Epoch 19/1000\n",
      "40/40 - 0s - loss: 0.2179 - 312ms/epoch - 8ms/step\n",
      "Epoch 20/1000\n",
      "40/40 - 0s - loss: 0.2064 - 313ms/epoch - 8ms/step\n",
      "Epoch 21/1000\n",
      "40/40 - 0s - loss: 0.1902 - 304ms/epoch - 8ms/step\n",
      "Epoch 22/1000\n",
      "40/40 - 0s - loss: 0.1676 - 306ms/epoch - 8ms/step\n",
      "Epoch 23/1000\n",
      "40/40 - 0s - loss: 0.1524 - 288ms/epoch - 7ms/step\n",
      "Epoch 24/1000\n",
      "40/40 - 0s - loss: 0.1333 - 308ms/epoch - 8ms/step\n",
      "Epoch 25/1000\n",
      "40/40 - 0s - loss: 0.1211 - 303ms/epoch - 8ms/step\n",
      "Epoch 26/1000\n",
      "40/40 - 0s - loss: 0.2013 - 299ms/epoch - 7ms/step\n",
      "Epoch 27/1000\n",
      "40/40 - 0s - loss: 0.1794 - 304ms/epoch - 8ms/step\n",
      "Epoch 28/1000\n",
      "40/40 - 0s - loss: 0.1320 - 293ms/epoch - 7ms/step\n",
      "Epoch 29/1000\n",
      "40/40 - 0s - loss: 0.1166 - 298ms/epoch - 7ms/step\n",
      "Epoch 30/1000\n",
      "40/40 - 0s - loss: 0.1127 - 290ms/epoch - 7ms/step\n",
      "Epoch 31/1000\n",
      "40/40 - 0s - loss: 0.1100 - 315ms/epoch - 8ms/step\n",
      "Epoch 32/1000\n",
      "40/40 - 0s - loss: 0.1076 - 292ms/epoch - 7ms/step\n",
      "Epoch 33/1000\n",
      "40/40 - 0s - loss: 0.1059 - 290ms/epoch - 7ms/step\n",
      "Epoch 34/1000\n",
      "40/40 - 0s - loss: 0.1053 - 291ms/epoch - 7ms/step\n",
      "Epoch 35/1000\n",
      "40/40 - 0s - loss: 0.1068 - 297ms/epoch - 7ms/step\n",
      "Epoch 36/1000\n",
      "40/40 - 0s - loss: 0.1044 - 311ms/epoch - 8ms/step\n",
      "Epoch 37/1000\n",
      "40/40 - 0s - loss: 0.1049 - 302ms/epoch - 8ms/step\n",
      "Epoch 38/1000\n",
      "40/40 - 0s - loss: 0.1047 - 294ms/epoch - 7ms/step\n",
      "Epoch 39/1000\n",
      "40/40 - 0s - loss: 0.1044 - 303ms/epoch - 8ms/step\n",
      "Epoch 40/1000\n",
      "40/40 - 0s - loss: 0.1041 - 316ms/epoch - 8ms/step\n",
      "Epoch 41/1000\n",
      "40/40 - 0s - loss: 0.1039 - 328ms/epoch - 8ms/step\n",
      "Epoch 42/1000\n",
      "40/40 - 0s - loss: 0.1040 - 314ms/epoch - 8ms/step\n",
      "Epoch 43/1000\n",
      "40/40 - 0s - loss: 0.1038 - 297ms/epoch - 7ms/step\n",
      "Epoch 44/1000\n",
      "40/40 - 0s - loss: 0.1037 - 318ms/epoch - 8ms/step\n",
      "Epoch 45/1000\n",
      "40/40 - 0s - loss: 0.1036 - 311ms/epoch - 8ms/step\n",
      "Epoch 46/1000\n",
      "40/40 - 0s - loss: 0.1036 - 313ms/epoch - 8ms/step\n",
      "Epoch 47/1000\n",
      "40/40 - 0s - loss: 0.1035 - 322ms/epoch - 8ms/step\n",
      "Epoch 48/1000\n",
      "40/40 - 0s - loss: 0.1037 - 305ms/epoch - 8ms/step\n",
      "Epoch 49/1000\n",
      "40/40 - 0s - loss: 0.1033 - 307ms/epoch - 8ms/step\n",
      "Epoch 50/1000\n",
      "40/40 - 0s - loss: 0.1031 - 316ms/epoch - 8ms/step\n",
      "Epoch 51/1000\n",
      "40/40 - 0s - loss: 0.1032 - 333ms/epoch - 8ms/step\n",
      "Epoch 52/1000\n",
      "40/40 - 0s - loss: 0.1029 - 314ms/epoch - 8ms/step\n",
      "Epoch 53/1000\n",
      "40/40 - 0s - loss: 0.1028 - 297ms/epoch - 7ms/step\n",
      "Epoch 54/1000\n",
      "40/40 - 0s - loss: 0.1026 - 333ms/epoch - 8ms/step\n",
      "Epoch 55/1000\n",
      "40/40 - 0s - loss: 0.1038 - 320ms/epoch - 8ms/step\n",
      "Epoch 56/1000\n",
      "40/40 - 0s - loss: 0.1035 - 302ms/epoch - 8ms/step\n",
      "Epoch 57/1000\n",
      "40/40 - 0s - loss: 0.1023 - 325ms/epoch - 8ms/step\n",
      "Epoch 58/1000\n",
      "40/40 - 0s - loss: 0.1011 - 310ms/epoch - 8ms/step\n",
      "Epoch 59/1000\n",
      "40/40 - 0s - loss: 0.1028 - 318ms/epoch - 8ms/step\n",
      "Epoch 60/1000\n",
      "40/40 - 0s - loss: 0.1027 - 310ms/epoch - 8ms/step\n",
      "Epoch 61/1000\n",
      "40/40 - 0s - loss: 0.1022 - 356ms/epoch - 9ms/step\n",
      "Epoch 62/1000\n",
      "40/40 - 0s - loss: 0.1015 - 353ms/epoch - 9ms/step\n",
      "Epoch 63/1000\n",
      "40/40 - 0s - loss: 0.1029 - 335ms/epoch - 8ms/step\n",
      "Epoch 64/1000\n",
      "40/40 - 0s - loss: 0.1031 - 359ms/epoch - 9ms/step\n",
      "Epoch 65/1000\n",
      "40/40 - 0s - loss: 0.1028 - 348ms/epoch - 9ms/step\n",
      "Epoch 66/1000\n",
      "40/40 - 0s - loss: 0.1025 - 344ms/epoch - 9ms/step\n",
      "Epoch 67/1000\n",
      "40/40 - 0s - loss: 0.1026 - 309ms/epoch - 8ms/step\n",
      "Epoch 68/1000\n",
      "40/40 - 0s - loss: 0.1024 - 318ms/epoch - 8ms/step\n",
      "Epoch 69/1000\n",
      "40/40 - 0s - loss: 0.1023 - 303ms/epoch - 8ms/step\n",
      "Epoch 70/1000\n",
      "40/40 - 0s - loss: 0.1010 - 329ms/epoch - 8ms/step\n",
      "Epoch 71/1000\n",
      "40/40 - 0s - loss: 0.0997 - 312ms/epoch - 8ms/step\n",
      "Epoch 72/1000\n",
      "40/40 - 0s - loss: 0.0996 - 338ms/epoch - 8ms/step\n",
      "Epoch 73/1000\n",
      "40/40 - 0s - loss: 0.1027 - 297ms/epoch - 7ms/step\n",
      "Epoch 74/1000\n",
      "40/40 - 0s - loss: 0.1018 - 328ms/epoch - 8ms/step\n",
      "Epoch 75/1000\n",
      "40/40 - 0s - loss: 0.1021 - 317ms/epoch - 8ms/step\n",
      "Epoch 76/1000\n",
      "40/40 - 0s - loss: 0.1015 - 315ms/epoch - 8ms/step\n",
      "Epoch 77/1000\n",
      "40/40 - 0s - loss: 0.0994 - 325ms/epoch - 8ms/step\n",
      "Epoch 78/1000\n",
      "40/40 - 0s - loss: 0.0978 - 305ms/epoch - 8ms/step\n",
      "Epoch 79/1000\n",
      "40/40 - 0s - loss: 0.0945 - 291ms/epoch - 7ms/step\n",
      "Epoch 80/1000\n",
      "40/40 - 0s - loss: 0.0979 - 339ms/epoch - 8ms/step\n",
      "Epoch 81/1000\n",
      "40/40 - 0s - loss: 0.0957 - 322ms/epoch - 8ms/step\n",
      "Epoch 82/1000\n",
      "40/40 - 0s - loss: 0.0952 - 314ms/epoch - 8ms/step\n",
      "Epoch 83/1000\n",
      "40/40 - 0s - loss: 0.0915 - 300ms/epoch - 7ms/step\n",
      "Epoch 84/1000\n",
      "40/40 - 0s - loss: 0.0918 - 312ms/epoch - 8ms/step\n",
      "Epoch 85/1000\n",
      "40/40 - 0s - loss: 0.0902 - 321ms/epoch - 8ms/step\n",
      "Epoch 86/1000\n",
      "40/40 - 0s - loss: 0.0913 - 321ms/epoch - 8ms/step\n",
      "Epoch 87/1000\n",
      "40/40 - 0s - loss: 0.0894 - 333ms/epoch - 8ms/step\n",
      "Epoch 88/1000\n",
      "40/40 - 0s - loss: 0.0875 - 335ms/epoch - 8ms/step\n",
      "Epoch 89/1000\n",
      "40/40 - 0s - loss: 0.0853 - 329ms/epoch - 8ms/step\n",
      "Epoch 90/1000\n",
      "40/40 - 0s - loss: 0.0837 - 364ms/epoch - 9ms/step\n",
      "Epoch 91/1000\n",
      "40/40 - 0s - loss: 0.0826 - 360ms/epoch - 9ms/step\n",
      "Epoch 92/1000\n",
      "40/40 - 0s - loss: 0.0816 - 333ms/epoch - 8ms/step\n",
      "Epoch 93/1000\n",
      "40/40 - 0s - loss: 0.0795 - 339ms/epoch - 8ms/step\n",
      "Epoch 94/1000\n",
      "40/40 - 0s - loss: 0.0801 - 320ms/epoch - 8ms/step\n",
      "Epoch 95/1000\n",
      "40/40 - 0s - loss: 0.0765 - 313ms/epoch - 8ms/step\n",
      "Epoch 96/1000\n",
      "40/40 - 0s - loss: 0.0735 - 323ms/epoch - 8ms/step\n",
      "Epoch 97/1000\n",
      "40/40 - 0s - loss: 0.0739 - 316ms/epoch - 8ms/step\n",
      "Epoch 98/1000\n",
      "40/40 - 0s - loss: 0.0707 - 314ms/epoch - 8ms/step\n",
      "Epoch 99/1000\n",
      "40/40 - 0s - loss: 0.0705 - 337ms/epoch - 8ms/step\n",
      "Epoch 100/1000\n",
      "40/40 - 0s - loss: 0.0723 - 318ms/epoch - 8ms/step\n",
      "Epoch 101/1000\n",
      "40/40 - 0s - loss: 0.0755 - 314ms/epoch - 8ms/step\n",
      "Epoch 102/1000\n",
      "40/40 - 0s - loss: 0.0713 - 324ms/epoch - 8ms/step\n",
      "Epoch 103/1000\n",
      "40/40 - 0s - loss: 0.0698 - 324ms/epoch - 8ms/step\n",
      "Epoch 104/1000\n",
      "40/40 - 0s - loss: 0.0667 - 325ms/epoch - 8ms/step\n",
      "Epoch 105/1000\n",
      "40/40 - 0s - loss: 0.0656 - 305ms/epoch - 8ms/step\n",
      "Epoch 106/1000\n",
      "40/40 - 0s - loss: 0.0657 - 301ms/epoch - 8ms/step\n",
      "Epoch 107/1000\n",
      "40/40 - 0s - loss: 0.0607 - 319ms/epoch - 8ms/step\n",
      "Epoch 108/1000\n",
      "40/40 - 0s - loss: 0.0598 - 330ms/epoch - 8ms/step\n",
      "Epoch 109/1000\n",
      "40/40 - 0s - loss: 0.0560 - 315ms/epoch - 8ms/step\n",
      "Epoch 110/1000\n",
      "40/40 - 0s - loss: 0.0521 - 346ms/epoch - 9ms/step\n",
      "Epoch 111/1000\n",
      "40/40 - 0s - loss: 0.0548 - 344ms/epoch - 9ms/step\n",
      "Epoch 112/1000\n",
      "40/40 - 0s - loss: 0.0655 - 365ms/epoch - 9ms/step\n",
      "Epoch 113/1000\n",
      "40/40 - 0s - loss: 0.0546 - 350ms/epoch - 9ms/step\n",
      "Epoch 114/1000\n",
      "40/40 - 0s - loss: 0.0623 - 406ms/epoch - 10ms/step\n",
      "Epoch 115/1000\n",
      "40/40 - 0s - loss: 0.0515 - 380ms/epoch - 9ms/step\n",
      "Epoch 116/1000\n",
      "40/40 - 0s - loss: 0.0498 - 325ms/epoch - 8ms/step\n",
      "Epoch 117/1000\n",
      "40/40 - 0s - loss: 0.0635 - 400ms/epoch - 10ms/step\n",
      "Epoch 118/1000\n",
      "40/40 - 0s - loss: 0.0507 - 331ms/epoch - 8ms/step\n",
      "Epoch 119/1000\n",
      "40/40 - 0s - loss: 0.0480 - 323ms/epoch - 8ms/step\n",
      "Epoch 120/1000\n",
      "40/40 - 0s - loss: 0.0439 - 366ms/epoch - 9ms/step\n",
      "Epoch 121/1000\n",
      "40/40 - 0s - loss: 0.0474 - 340ms/epoch - 9ms/step\n",
      "Epoch 122/1000\n",
      "40/40 - 0s - loss: 0.0450 - 354ms/epoch - 9ms/step\n",
      "Epoch 123/1000\n",
      "40/40 - 0s - loss: 0.0498 - 344ms/epoch - 9ms/step\n",
      "Epoch 124/1000\n",
      "40/40 - 0s - loss: 0.0447 - 308ms/epoch - 8ms/step\n",
      "Epoch 125/1000\n",
      "40/40 - 0s - loss: 0.0477 - 323ms/epoch - 8ms/step\n",
      "Epoch 126/1000\n",
      "40/40 - 0s - loss: 0.0452 - 353ms/epoch - 9ms/step\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 127/1000\n",
      "40/40 - 0s - loss: 0.0451 - 330ms/epoch - 8ms/step\n",
      "Epoch 128/1000\n",
      "40/40 - 0s - loss: 0.0687 - 327ms/epoch - 8ms/step\n",
      "Epoch 129/1000\n",
      "40/40 - 0s - loss: 0.1097 - 309ms/epoch - 8ms/step\n",
      "Epoch 130/1000\n",
      "40/40 - 0s - loss: 0.0901 - 324ms/epoch - 8ms/step\n",
      "Epoch 131/1000\n",
      "40/40 - 0s - loss: 0.0512 - 306ms/epoch - 8ms/step\n",
      "Epoch 132/1000\n",
      "40/40 - 0s - loss: 0.0468 - 311ms/epoch - 8ms/step\n",
      "Epoch 133/1000\n",
      "40/40 - 0s - loss: 0.0631 - 312ms/epoch - 8ms/step\n",
      "Epoch 134/1000\n",
      "40/40 - 0s - loss: 0.0509 - 306ms/epoch - 8ms/step\n",
      "Epoch 135/1000\n",
      "40/40 - 0s - loss: 0.0469 - 333ms/epoch - 8ms/step\n",
      "Epoch 136/1000\n",
      "40/40 - 0s - loss: 0.0436 - 322ms/epoch - 8ms/step\n",
      "Epoch 137/1000\n",
      "40/40 - 0s - loss: 0.0470 - 321ms/epoch - 8ms/step\n",
      "Epoch 138/1000\n",
      "40/40 - 0s - loss: 0.0394 - 325ms/epoch - 8ms/step\n",
      "Epoch 139/1000\n",
      "40/40 - 0s - loss: 0.0412 - 338ms/epoch - 8ms/step\n",
      "Epoch 140/1000\n",
      "40/40 - 0s - loss: 0.0426 - 304ms/epoch - 8ms/step\n",
      "Epoch 141/1000\n",
      "40/40 - 0s - loss: 0.0409 - 337ms/epoch - 8ms/step\n",
      "Epoch 142/1000\n",
      "40/40 - 0s - loss: 0.0413 - 364ms/epoch - 9ms/step\n",
      "Epoch 143/1000\n",
      "40/40 - 0s - loss: 0.0376 - 311ms/epoch - 8ms/step\n",
      "Epoch 144/1000\n",
      "40/40 - 0s - loss: 0.0390 - 341ms/epoch - 9ms/step\n",
      "Epoch 145/1000\n",
      "40/40 - 0s - loss: 0.0378 - 341ms/epoch - 9ms/step\n",
      "Epoch 146/1000\n",
      "40/40 - 0s - loss: 0.0439 - 323ms/epoch - 8ms/step\n",
      "Epoch 147/1000\n",
      "40/40 - 0s - loss: 0.0405 - 357ms/epoch - 9ms/step\n",
      "Epoch 148/1000\n",
      "40/40 - 0s - loss: 0.0394 - 351ms/epoch - 9ms/step\n",
      "Epoch 149/1000\n",
      "40/40 - 0s - loss: 0.0366 - 349ms/epoch - 9ms/step\n",
      "Epoch 150/1000\n",
      "40/40 - 0s - loss: 0.0421 - 333ms/epoch - 8ms/step\n",
      "Epoch 151/1000\n",
      "40/40 - 0s - loss: 0.0414 - 333ms/epoch - 8ms/step\n",
      "Epoch 152/1000\n",
      "40/40 - 0s - loss: 0.0397 - 401ms/epoch - 10ms/step\n",
      "Epoch 153/1000\n",
      "40/40 - 0s - loss: 0.0401 - 375ms/epoch - 9ms/step\n",
      "Epoch 154/1000\n",
      "40/40 - 0s - loss: 0.0431 - 407ms/epoch - 10ms/step\n",
      "Epoch 155/1000\n",
      "40/40 - 0s - loss: 0.0466 - 364ms/epoch - 9ms/step\n",
      "Epoch 156/1000\n",
      "40/40 - 0s - loss: 0.0395 - 328ms/epoch - 8ms/step\n",
      "Epoch 157/1000\n",
      "40/40 - 0s - loss: 0.0382 - 356ms/epoch - 9ms/step\n",
      "Epoch 158/1000\n",
      "40/40 - 0s - loss: 0.0394 - 362ms/epoch - 9ms/step\n",
      "Epoch 159/1000\n",
      "40/40 - 0s - loss: 0.0545 - 365ms/epoch - 9ms/step\n",
      "Epoch 160/1000\n",
      "40/40 - 0s - loss: 0.0524 - 382ms/epoch - 10ms/step\n",
      "Epoch 161/1000\n",
      "40/40 - 0s - loss: 0.0496 - 368ms/epoch - 9ms/step\n",
      "Epoch 162/1000\n",
      "40/40 - 0s - loss: 0.0478 - 345ms/epoch - 9ms/step\n",
      "Epoch 163/1000\n",
      "40/40 - 0s - loss: 0.0396 - 329ms/epoch - 8ms/step\n",
      "Epoch 164/1000\n",
      "40/40 - 0s - loss: 0.0425 - 307ms/epoch - 8ms/step\n",
      "Epoch 165/1000\n",
      "40/40 - 0s - loss: 0.0425 - 326ms/epoch - 8ms/step\n",
      "Epoch 166/1000\n",
      "40/40 - 0s - loss: 0.0398 - 323ms/epoch - 8ms/step\n",
      "Epoch 167/1000\n",
      "40/40 - 0s - loss: 0.0399 - 319ms/epoch - 8ms/step\n",
      "Epoch 168/1000\n",
      "40/40 - 0s - loss: 0.0383 - 324ms/epoch - 8ms/step\n",
      "Epoch 169/1000\n",
      "40/40 - 0s - loss: 0.0395 - 308ms/epoch - 8ms/step\n",
      "Epoch 170/1000\n",
      "40/40 - 0s - loss: 0.0383 - 328ms/epoch - 8ms/step\n",
      "Epoch 171/1000\n",
      "40/40 - 0s - loss: 0.0371 - 309ms/epoch - 8ms/step\n",
      "Epoch 172/1000\n",
      "40/40 - 0s - loss: 0.0414 - 341ms/epoch - 9ms/step\n",
      "Epoch 173/1000\n",
      "40/40 - 0s - loss: 0.0385 - 323ms/epoch - 8ms/step\n",
      "Epoch 174/1000\n",
      "40/40 - 0s - loss: 0.0464 - 312ms/epoch - 8ms/step\n",
      "Epoch 175/1000\n",
      "40/40 - 0s - loss: 0.0423 - 325ms/epoch - 8ms/step\n",
      "Epoch 176/1000\n",
      "40/40 - 0s - loss: 0.0370 - 322ms/epoch - 8ms/step\n",
      "Epoch 177/1000\n",
      "40/40 - 0s - loss: 0.0417 - 325ms/epoch - 8ms/step\n",
      "Epoch 178/1000\n",
      "40/40 - 0s - loss: 0.0365 - 297ms/epoch - 7ms/step\n",
      "Epoch 179/1000\n",
      "40/40 - 0s - loss: 0.0408 - 324ms/epoch - 8ms/step\n",
      "Epoch 180/1000\n",
      "40/40 - 0s - loss: 0.0401 - 322ms/epoch - 8ms/step\n",
      "Epoch 181/1000\n",
      "40/40 - 0s - loss: 0.0360 - 307ms/epoch - 8ms/step\n",
      "Epoch 182/1000\n",
      "40/40 - 0s - loss: 0.0354 - 356ms/epoch - 9ms/step\n",
      "Epoch 183/1000\n",
      "40/40 - 0s - loss: 0.0375 - 329ms/epoch - 8ms/step\n",
      "Epoch 184/1000\n",
      "40/40 - 0s - loss: 0.0355 - 305ms/epoch - 8ms/step\n",
      "Epoch 185/1000\n",
      "40/40 - 0s - loss: 0.0368 - 312ms/epoch - 8ms/step\n",
      "Epoch 186/1000\n",
      "40/40 - 0s - loss: 0.0393 - 313ms/epoch - 8ms/step\n",
      "Epoch 187/1000\n",
      "40/40 - 0s - loss: 0.0389 - 307ms/epoch - 8ms/step\n",
      "Epoch 188/1000\n",
      "40/40 - 0s - loss: 0.0385 - 329ms/epoch - 8ms/step\n",
      "Epoch 189/1000\n",
      "40/40 - 0s - loss: 0.0341 - 296ms/epoch - 7ms/step\n",
      "Epoch 190/1000\n",
      "40/40 - 0s - loss: 0.0373 - 319ms/epoch - 8ms/step\n",
      "Epoch 191/1000\n",
      "40/40 - 0s - loss: 0.0366 - 321ms/epoch - 8ms/step\n",
      "Epoch 192/1000\n",
      "40/40 - 0s - loss: 0.0449 - 338ms/epoch - 8ms/step\n",
      "Epoch 193/1000\n",
      "40/40 - 0s - loss: 0.0395 - 319ms/epoch - 8ms/step\n",
      "Epoch 194/1000\n",
      "40/40 - 0s - loss: 0.0468 - 314ms/epoch - 8ms/step\n",
      "Epoch 195/1000\n",
      "40/40 - 0s - loss: 0.0414 - 304ms/epoch - 8ms/step\n",
      "Epoch 196/1000\n",
      "40/40 - 0s - loss: 0.0433 - 312ms/epoch - 8ms/step\n",
      "Epoch 197/1000\n",
      "40/40 - 0s - loss: 0.0362 - 326ms/epoch - 8ms/step\n",
      "Epoch 198/1000\n",
      "40/40 - 0s - loss: 0.0342 - 325ms/epoch - 8ms/step\n",
      "Epoch 199/1000\n",
      "40/40 - 0s - loss: 0.0336 - 384ms/epoch - 10ms/step\n",
      "Epoch 200/1000\n",
      "40/40 - 0s - loss: 0.0331 - 405ms/epoch - 10ms/step\n",
      "Epoch 201/1000\n",
      "40/40 - 0s - loss: 0.0342 - 379ms/epoch - 9ms/step\n",
      "Epoch 202/1000\n",
      "40/40 - 0s - loss: 0.0345 - 370ms/epoch - 9ms/step\n",
      "Epoch 203/1000\n",
      "40/40 - 0s - loss: 0.0425 - 327ms/epoch - 8ms/step\n",
      "Epoch 204/1000\n",
      "40/40 - 0s - loss: 0.0450 - 323ms/epoch - 8ms/step\n",
      "Epoch 205/1000\n",
      "40/40 - 0s - loss: 0.0373 - 368ms/epoch - 9ms/step\n",
      "Epoch 206/1000\n",
      "40/40 - 0s - loss: 0.0328 - 368ms/epoch - 9ms/step\n",
      "Epoch 207/1000\n",
      "40/40 - 0s - loss: 0.0322 - 353ms/epoch - 9ms/step\n",
      "Epoch 208/1000\n",
      "40/40 - 0s - loss: 0.0396 - 383ms/epoch - 10ms/step\n",
      "Epoch 209/1000\n",
      "40/40 - 0s - loss: 0.0352 - 373ms/epoch - 9ms/step\n",
      "Epoch 210/1000\n",
      "40/40 - 0s - loss: 0.0316 - 346ms/epoch - 9ms/step\n",
      "Epoch 211/1000\n",
      "40/40 - 0s - loss: 0.0358 - 331ms/epoch - 8ms/step\n",
      "Epoch 212/1000\n",
      "40/40 - 0s - loss: 0.0365 - 320ms/epoch - 8ms/step\n",
      "Epoch 213/1000\n",
      "40/40 - 0s - loss: 0.0343 - 337ms/epoch - 8ms/step\n",
      "Epoch 214/1000\n",
      "40/40 - 0s - loss: 0.0316 - 320ms/epoch - 8ms/step\n",
      "Epoch 215/1000\n",
      "40/40 - 0s - loss: 0.0331 - 333ms/epoch - 8ms/step\n",
      "Epoch 216/1000\n",
      "40/40 - 0s - loss: 0.0333 - 322ms/epoch - 8ms/step\n",
      "Epoch 217/1000\n",
      "40/40 - 0s - loss: 0.0333 - 331ms/epoch - 8ms/step\n",
      "Epoch 218/1000\n",
      "40/40 - 0s - loss: 0.0397 - 327ms/epoch - 8ms/step\n",
      "Epoch 219/1000\n",
      "40/40 - 0s - loss: 0.0418 - 348ms/epoch - 9ms/step\n",
      "Epoch 220/1000\n",
      "40/40 - 0s - loss: 0.0352 - 326ms/epoch - 8ms/step\n",
      "Epoch 221/1000\n",
      "40/40 - 0s - loss: 0.0366 - 318ms/epoch - 8ms/step\n",
      "Epoch 222/1000\n",
      "40/40 - 0s - loss: 0.0357 - 319ms/epoch - 8ms/step\n",
      "Epoch 223/1000\n",
      "40/40 - 0s - loss: 0.0317 - 323ms/epoch - 8ms/step\n",
      "Epoch 224/1000\n",
      "40/40 - 0s - loss: 0.0316 - 329ms/epoch - 8ms/step\n",
      "Epoch 225/1000\n",
      "40/40 - 0s - loss: 0.0296 - 329ms/epoch - 8ms/step\n",
      "Epoch 226/1000\n",
      "40/40 - 0s - loss: 0.0307 - 335ms/epoch - 8ms/step\n",
      "Epoch 227/1000\n",
      "40/40 - 0s - loss: 0.0362 - 326ms/epoch - 8ms/step\n",
      "Epoch 228/1000\n",
      "40/40 - 0s - loss: 0.0492 - 348ms/epoch - 9ms/step\n",
      "Epoch 229/1000\n",
      "40/40 - 0s - loss: 0.0340 - 329ms/epoch - 8ms/step\n",
      "Epoch 230/1000\n",
      "40/40 - 0s - loss: 0.0320 - 309ms/epoch - 8ms/step\n",
      "Epoch 231/1000\n",
      "40/40 - 0s - loss: 0.0287 - 318ms/epoch - 8ms/step\n",
      "Epoch 232/1000\n",
      "40/40 - 0s - loss: 0.0302 - 328ms/epoch - 8ms/step\n",
      "Epoch 233/1000\n",
      "40/40 - 0s - loss: 0.0292 - 342ms/epoch - 9ms/step\n",
      "Epoch 234/1000\n",
      "40/40 - 0s - loss: 0.0306 - 326ms/epoch - 8ms/step\n",
      "Epoch 235/1000\n",
      "40/40 - 0s - loss: 0.0271 - 309ms/epoch - 8ms/step\n",
      "Epoch 236/1000\n",
      "40/40 - 0s - loss: 0.0410 - 342ms/epoch - 9ms/step\n",
      "Epoch 237/1000\n",
      "40/40 - 0s - loss: 0.0333 - 307ms/epoch - 8ms/step\n",
      "Epoch 238/1000\n",
      "40/40 - 0s - loss: 0.0282 - 322ms/epoch - 8ms/step\n",
      "Epoch 239/1000\n",
      "40/40 - 0s - loss: 0.0277 - 320ms/epoch - 8ms/step\n",
      "Epoch 240/1000\n",
      "40/40 - 0s - loss: 0.0353 - 314ms/epoch - 8ms/step\n",
      "Epoch 241/1000\n",
      "40/40 - 0s - loss: 0.0313 - 324ms/epoch - 8ms/step\n",
      "Epoch 242/1000\n",
      "40/40 - 0s - loss: 0.0269 - 323ms/epoch - 8ms/step\n",
      "Epoch 243/1000\n",
      "40/40 - 0s - loss: 0.0294 - 336ms/epoch - 8ms/step\n",
      "Epoch 244/1000\n",
      "40/40 - 0s - loss: 0.0287 - 315ms/epoch - 8ms/step\n",
      "Epoch 245/1000\n",
      "40/40 - 0s - loss: 0.0285 - 332ms/epoch - 8ms/step\n",
      "Epoch 246/1000\n",
      "40/40 - 0s - loss: 0.0276 - 324ms/epoch - 8ms/step\n",
      "Epoch 247/1000\n",
      "40/40 - 0s - loss: 0.0248 - 319ms/epoch - 8ms/step\n",
      "Epoch 248/1000\n",
      "40/40 - 0s - loss: 0.0276 - 331ms/epoch - 8ms/step\n",
      "Epoch 249/1000\n",
      "40/40 - 0s - loss: 0.0280 - 314ms/epoch - 8ms/step\n",
      "Epoch 250/1000\n",
      "40/40 - 0s - loss: 0.0248 - 327ms/epoch - 8ms/step\n",
      "Epoch 251/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "40/40 - 0s - loss: 0.0263 - 319ms/epoch - 8ms/step\n",
      "Epoch 252/1000\n",
      "40/40 - 0s - loss: 0.0492 - 342ms/epoch - 9ms/step\n",
      "Epoch 253/1000\n",
      "40/40 - 0s - loss: 0.0515 - 370ms/epoch - 9ms/step\n",
      "Epoch 254/1000\n",
      "40/40 - 0s - loss: 0.0377 - 353ms/epoch - 9ms/step\n",
      "Epoch 255/1000\n",
      "40/40 - 0s - loss: 0.0336 - 364ms/epoch - 9ms/step\n",
      "Epoch 256/1000\n",
      "40/40 - 0s - loss: 0.0344 - 386ms/epoch - 10ms/step\n",
      "Epoch 257/1000\n",
      "40/40 - 0s - loss: 0.0355 - 364ms/epoch - 9ms/step\n",
      "Epoch 258/1000\n",
      "40/40 - 0s - loss: 0.0346 - 342ms/epoch - 9ms/step\n",
      "Epoch 259/1000\n",
      "40/40 - 0s - loss: 0.0331 - 325ms/epoch - 8ms/step\n",
      "Epoch 260/1000\n",
      "40/40 - 0s - loss: 0.0338 - 323ms/epoch - 8ms/step\n",
      "Epoch 261/1000\n",
      "40/40 - 0s - loss: 0.0320 - 326ms/epoch - 8ms/step\n",
      "Epoch 262/1000\n",
      "40/40 - 0s - loss: 0.0336 - 323ms/epoch - 8ms/step\n",
      "Epoch 263/1000\n",
      "40/40 - 0s - loss: 0.0343 - 331ms/epoch - 8ms/step\n",
      "Epoch 264/1000\n",
      "40/40 - 0s - loss: 0.0408 - 321ms/epoch - 8ms/step\n",
      "Epoch 265/1000\n",
      "40/40 - 0s - loss: 0.0338 - 332ms/epoch - 8ms/step\n",
      "Epoch 266/1000\n",
      "40/40 - 0s - loss: 0.0341 - 347ms/epoch - 9ms/step\n",
      "Epoch 267/1000\n",
      "40/40 - 0s - loss: 0.0319 - 316ms/epoch - 8ms/step\n",
      "Epoch 268/1000\n",
      "40/40 - 0s - loss: 0.0299 - 338ms/epoch - 8ms/step\n",
      "Epoch 269/1000\n",
      "40/40 - 0s - loss: 0.0403 - 325ms/epoch - 8ms/step\n",
      "Epoch 270/1000\n",
      "40/40 - 0s - loss: 0.0329 - 322ms/epoch - 8ms/step\n",
      "Epoch 271/1000\n",
      "40/40 - 0s - loss: 0.0342 - 328ms/epoch - 8ms/step\n",
      "Epoch 272/1000\n",
      "40/40 - 0s - loss: 0.0304 - 367ms/epoch - 9ms/step\n",
      "Epoch 273/1000\n",
      "40/40 - 0s - loss: 0.0371 - 386ms/epoch - 10ms/step\n",
      "Epoch 274/1000\n",
      "40/40 - 0s - loss: 0.0350 - 396ms/epoch - 10ms/step\n",
      "Epoch 275/1000\n",
      "40/40 - 0s - loss: 0.0330 - 371ms/epoch - 9ms/step\n",
      "Epoch 276/1000\n",
      "40/40 - 0s - loss: 0.0311 - 344ms/epoch - 9ms/step\n",
      "Epoch 277/1000\n",
      "40/40 - 0s - loss: 0.0282 - 332ms/epoch - 8ms/step\n",
      "Epoch 278/1000\n",
      "40/40 - 0s - loss: 0.0280 - 385ms/epoch - 10ms/step\n",
      "Epoch 279/1000\n",
      "40/40 - 0s - loss: 0.0273 - 342ms/epoch - 9ms/step\n",
      "Epoch 280/1000\n",
      "40/40 - 0s - loss: 0.0336 - 331ms/epoch - 8ms/step\n",
      "Epoch 281/1000\n",
      "40/40 - 0s - loss: 0.0404 - 340ms/epoch - 8ms/step\n",
      "Epoch 282/1000\n",
      "40/40 - 0s - loss: 0.0317 - 366ms/epoch - 9ms/step\n",
      "Epoch 283/1000\n",
      "40/40 - 0s - loss: 0.0265 - 383ms/epoch - 10ms/step\n",
      "Epoch 284/1000\n",
      "40/40 - 0s - loss: 0.0264 - 326ms/epoch - 8ms/step\n",
      "Epoch 285/1000\n",
      "40/40 - 0s - loss: 0.0260 - 333ms/epoch - 8ms/step\n",
      "Epoch 286/1000\n",
      "40/40 - 0s - loss: 0.0354 - 342ms/epoch - 9ms/step\n",
      "Epoch 287/1000\n",
      "40/40 - 0s - loss: 0.0334 - 350ms/epoch - 9ms/step\n",
      "Epoch 288/1000\n",
      "40/40 - 0s - loss: 0.0307 - 335ms/epoch - 8ms/step\n",
      "Epoch 289/1000\n",
      "40/40 - 0s - loss: 0.0292 - 358ms/epoch - 9ms/step\n",
      "Epoch 290/1000\n",
      "40/40 - 0s - loss: 0.0283 - 378ms/epoch - 9ms/step\n",
      "Epoch 291/1000\n",
      "40/40 - 0s - loss: 0.0264 - 348ms/epoch - 9ms/step\n",
      "Epoch 292/1000\n",
      "40/40 - 0s - loss: 0.0247 - 372ms/epoch - 9ms/step\n",
      "Epoch 293/1000\n",
      "40/40 - 0s - loss: 0.0427 - 333ms/epoch - 8ms/step\n",
      "Epoch 294/1000\n",
      "40/40 - 0s - loss: 0.0401 - 323ms/epoch - 8ms/step\n",
      "Epoch 295/1000\n",
      "40/40 - 0s - loss: 0.0274 - 321ms/epoch - 8ms/step\n",
      "Epoch 296/1000\n",
      "40/40 - 0s - loss: 0.0246 - 343ms/epoch - 9ms/step\n",
      "Epoch 297/1000\n",
      "40/40 - 0s - loss: 0.0345 - 376ms/epoch - 9ms/step\n",
      "Epoch 298/1000\n",
      "40/40 - 0s - loss: 0.0292 - 325ms/epoch - 8ms/step\n",
      "Epoch 299/1000\n",
      "40/40 - 0s - loss: 0.0243 - 394ms/epoch - 10ms/step\n",
      "Epoch 300/1000\n",
      "40/40 - 0s - loss: 0.0231 - 424ms/epoch - 11ms/step\n",
      "Epoch 301/1000\n",
      "40/40 - 0s - loss: 0.0235 - 376ms/epoch - 9ms/step\n",
      "Epoch 302/1000\n",
      "40/40 - 0s - loss: 0.0432 - 402ms/epoch - 10ms/step\n",
      "Epoch 303/1000\n",
      "40/40 - 0s - loss: 0.0484 - 387ms/epoch - 10ms/step\n",
      "Epoch 304/1000\n",
      "40/40 - 0s - loss: 0.0375 - 372ms/epoch - 9ms/step\n",
      "Epoch 305/1000\n",
      "40/40 - 0s - loss: 0.0349 - 315ms/epoch - 8ms/step\n",
      "Epoch 306/1000\n",
      "40/40 - 0s - loss: 0.0336 - 365ms/epoch - 9ms/step\n",
      "Epoch 307/1000\n",
      "40/40 - 0s - loss: 0.0324 - 347ms/epoch - 9ms/step\n",
      "Epoch 308/1000\n",
      "40/40 - 0s - loss: 0.0332 - 326ms/epoch - 8ms/step\n",
      "Epoch 309/1000\n",
      "40/40 - 0s - loss: 0.0312 - 340ms/epoch - 9ms/step\n",
      "Epoch 310/1000\n",
      "40/40 - 0s - loss: 0.0306 - 350ms/epoch - 9ms/step\n",
      "Epoch 311/1000\n",
      "40/40 - 0s - loss: 0.0293 - 353ms/epoch - 9ms/step\n",
      "Epoch 312/1000\n",
      "40/40 - 0s - loss: 0.0302 - 391ms/epoch - 10ms/step\n",
      "Epoch 313/1000\n",
      "40/40 - 0s - loss: 0.0307 - 339ms/epoch - 8ms/step\n",
      "Epoch 314/1000\n",
      "40/40 - 0s - loss: 0.0300 - 332ms/epoch - 8ms/step\n",
      "Epoch 315/1000\n",
      "40/40 - 0s - loss: 0.0309 - 364ms/epoch - 9ms/step\n",
      "Epoch 316/1000\n",
      "40/40 - 0s - loss: 0.0386 - 401ms/epoch - 10ms/step\n",
      "Epoch 317/1000\n",
      "40/40 - 0s - loss: 0.0302 - 369ms/epoch - 9ms/step\n",
      "Epoch 318/1000\n",
      "40/40 - 0s - loss: 0.0313 - 341ms/epoch - 9ms/step\n",
      "Epoch 319/1000\n",
      "40/40 - 0s - loss: 0.0268 - 333ms/epoch - 8ms/step\n",
      "Epoch 320/1000\n",
      "40/40 - 0s - loss: 0.0268 - 339ms/epoch - 8ms/step\n",
      "Epoch 321/1000\n",
      "40/40 - 0s - loss: 0.0262 - 356ms/epoch - 9ms/step\n",
      "Epoch 322/1000\n",
      "40/40 - 0s - loss: 0.0248 - 319ms/epoch - 8ms/step\n",
      "Epoch 323/1000\n",
      "40/40 - 0s - loss: 0.0265 - 333ms/epoch - 8ms/step\n",
      "Epoch 324/1000\n",
      "40/40 - 0s - loss: 0.0257 - 346ms/epoch - 9ms/step\n",
      "Epoch 325/1000\n",
      "40/40 - 0s - loss: 0.0264 - 329ms/epoch - 8ms/step\n",
      "Epoch 326/1000\n",
      "40/40 - 0s - loss: 0.0273 - 341ms/epoch - 9ms/step\n",
      "Epoch 327/1000\n",
      "40/40 - 0s - loss: 0.0288 - 334ms/epoch - 8ms/step\n",
      "Epoch 328/1000\n",
      "40/40 - 0s - loss: 0.0314 - 470ms/epoch - 12ms/step\n",
      "Epoch 329/1000\n",
      "40/40 - 0s - loss: 0.0348 - 488ms/epoch - 12ms/step\n",
      "Epoch 330/1000\n",
      "40/40 - 0s - loss: 0.0582 - 386ms/epoch - 10ms/step\n",
      "Epoch 331/1000\n",
      "40/40 - 0s - loss: 0.0365 - 411ms/epoch - 10ms/step\n",
      "Epoch 332/1000\n",
      "40/40 - 0s - loss: 0.0346 - 450ms/epoch - 11ms/step\n",
      "Epoch 333/1000\n",
      "40/40 - 0s - loss: 0.0284 - 444ms/epoch - 11ms/step\n",
      "Epoch 334/1000\n",
      "40/40 - 0s - loss: 0.0374 - 431ms/epoch - 11ms/step\n",
      "Epoch 335/1000\n",
      "40/40 - 0s - loss: 0.0318 - 487ms/epoch - 12ms/step\n",
      "Epoch 336/1000\n",
      "40/40 - 0s - loss: 0.0291 - 411ms/epoch - 10ms/step\n",
      "Epoch 337/1000\n",
      "40/40 - 0s - loss: 0.0207 - 453ms/epoch - 11ms/step\n",
      "Epoch 338/1000\n",
      "40/40 - 0s - loss: 0.0266 - 446ms/epoch - 11ms/step\n",
      "Epoch 339/1000\n",
      "40/40 - 0s - loss: 0.0254 - 468ms/epoch - 12ms/step\n",
      "Epoch 340/1000\n",
      "40/40 - 0s - loss: 0.0269 - 418ms/epoch - 10ms/step\n",
      "Epoch 341/1000\n",
      "40/40 - 0s - loss: 0.0377 - 453ms/epoch - 11ms/step\n",
      "Epoch 342/1000\n",
      "40/40 - 0s - loss: 0.0258 - 442ms/epoch - 11ms/step\n",
      "Epoch 343/1000\n",
      "40/40 - 0s - loss: 0.0216 - 360ms/epoch - 9ms/step\n",
      "Epoch 344/1000\n",
      "40/40 - 0s - loss: 0.0233 - 464ms/epoch - 12ms/step\n",
      "Epoch 345/1000\n",
      "40/40 - 0s - loss: 0.0198 - 391ms/epoch - 10ms/step\n",
      "Epoch 346/1000\n",
      "40/40 - 0s - loss: 0.0215 - 337ms/epoch - 8ms/step\n",
      "Epoch 347/1000\n",
      "40/40 - 0s - loss: 0.0205 - 318ms/epoch - 8ms/step\n",
      "Epoch 348/1000\n",
      "40/40 - 0s - loss: 0.0219 - 322ms/epoch - 8ms/step\n",
      "Epoch 349/1000\n",
      "40/40 - 0s - loss: 0.0125 - 311ms/epoch - 8ms/step\n",
      "Epoch 350/1000\n",
      "40/40 - 0s - loss: 0.0088 - 392ms/epoch - 10ms/step\n",
      "Epoch 351/1000\n",
      "40/40 - 0s - loss: 0.0066 - 354ms/epoch - 9ms/step\n",
      "Epoch 352/1000\n",
      "40/40 - 0s - loss: 0.0150 - 311ms/epoch - 8ms/step\n",
      "Epoch 353/1000\n",
      "40/40 - 0s - loss: 0.0217 - 314ms/epoch - 8ms/step\n",
      "Epoch 354/1000\n",
      "40/40 - 0s - loss: 0.0160 - 306ms/epoch - 8ms/step\n",
      "Epoch 355/1000\n",
      "40/40 - 0s - loss: 0.0081 - 310ms/epoch - 8ms/step\n",
      "Epoch 356/1000\n",
      "40/40 - 0s - loss: 0.0079 - 354ms/epoch - 9ms/step\n",
      "Epoch 357/1000\n",
      "40/40 - 0s - loss: 0.0068 - 352ms/epoch - 9ms/step\n",
      "Epoch 358/1000\n",
      "40/40 - 0s - loss: 0.0066 - 341ms/epoch - 9ms/step\n",
      "Epoch 359/1000\n",
      "40/40 - 0s - loss: 0.0051 - 320ms/epoch - 8ms/step\n",
      "Epoch 360/1000\n",
      "40/40 - 0s - loss: 0.0049 - 323ms/epoch - 8ms/step\n",
      "Epoch 361/1000\n",
      "40/40 - 0s - loss: 0.0046 - 306ms/epoch - 8ms/step\n",
      "Epoch 362/1000\n",
      "40/40 - 0s - loss: 0.0048 - 334ms/epoch - 8ms/step\n",
      "Epoch 363/1000\n",
      "40/40 - 0s - loss: 0.0045 - 376ms/epoch - 9ms/step\n",
      "Epoch 364/1000\n",
      "40/40 - 0s - loss: 0.0046 - 362ms/epoch - 9ms/step\n",
      "Epoch 365/1000\n",
      "40/40 - 0s - loss: 0.0042 - 325ms/epoch - 8ms/step\n",
      "Epoch 366/1000\n",
      "40/40 - 0s - loss: 0.0040 - 341ms/epoch - 9ms/step\n",
      "Epoch 367/1000\n",
      "40/40 - 0s - loss: 0.0071 - 440ms/epoch - 11ms/step\n",
      "Epoch 368/1000\n",
      "40/40 - 0s - loss: 0.0046 - 451ms/epoch - 11ms/step\n",
      "Epoch 369/1000\n",
      "40/40 - 0s - loss: 0.0048 - 403ms/epoch - 10ms/step\n",
      "Epoch 370/1000\n",
      "40/40 - 0s - loss: 0.0078 - 363ms/epoch - 9ms/step\n",
      "Epoch 371/1000\n",
      "40/40 - 0s - loss: 0.0050 - 430ms/epoch - 11ms/step\n",
      "Epoch 372/1000\n",
      "40/40 - 0s - loss: 0.0040 - 384ms/epoch - 10ms/step\n",
      "Epoch 373/1000\n",
      "40/40 - 0s - loss: 0.0043 - 339ms/epoch - 8ms/step\n",
      "Epoch 374/1000\n",
      "40/40 - 0s - loss: 0.0171 - 369ms/epoch - 9ms/step\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 375/1000\n",
      "40/40 - 0s - loss: 0.0122 - 349ms/epoch - 9ms/step\n",
      "Epoch 376/1000\n",
      "40/40 - 0s - loss: 0.0135 - 404ms/epoch - 10ms/step\n",
      "Epoch 377/1000\n",
      "40/40 - 0s - loss: 0.0069 - 344ms/epoch - 9ms/step\n",
      "Epoch 378/1000\n",
      "40/40 - 0s - loss: 0.0066 - 326ms/epoch - 8ms/step\n",
      "Epoch 379/1000\n",
      "40/40 - 0s - loss: 0.0061 - 330ms/epoch - 8ms/step\n",
      "Epoch 380/1000\n",
      "40/40 - 0s - loss: 0.0058 - 319ms/epoch - 8ms/step\n",
      "Epoch 381/1000\n",
      "40/40 - 0s - loss: 0.0135 - 317ms/epoch - 8ms/step\n",
      "Epoch 382/1000\n",
      "40/40 - 0s - loss: 0.0046 - 310ms/epoch - 8ms/step\n",
      "Epoch 383/1000\n",
      "40/40 - 0s - loss: 0.0035 - 341ms/epoch - 9ms/step\n",
      "Epoch 384/1000\n",
      "40/40 - 0s - loss: 0.0028 - 354ms/epoch - 9ms/step\n",
      "Epoch 385/1000\n",
      "40/40 - 0s - loss: 0.0036 - 429ms/epoch - 11ms/step\n",
      "Epoch 386/1000\n",
      "40/40 - 0s - loss: 0.0028 - 407ms/epoch - 10ms/step\n",
      "Epoch 387/1000\n",
      "40/40 - 0s - loss: 0.0028 - 372ms/epoch - 9ms/step\n",
      "Epoch 388/1000\n",
      "40/40 - 0s - loss: 0.0044 - 355ms/epoch - 9ms/step\n",
      "Epoch 389/1000\n",
      "40/40 - 0s - loss: 0.0083 - 335ms/epoch - 8ms/step\n",
      "Epoch 390/1000\n",
      "40/40 - 0s - loss: 0.0060 - 337ms/epoch - 8ms/step\n",
      "Epoch 391/1000\n",
      "40/40 - 0s - loss: 0.0065 - 312ms/epoch - 8ms/step\n",
      "Epoch 392/1000\n",
      "40/40 - 0s - loss: 0.0039 - 327ms/epoch - 8ms/step\n",
      "Epoch 393/1000\n",
      "40/40 - 0s - loss: 0.0032 - 370ms/epoch - 9ms/step\n",
      "Epoch 394/1000\n",
      "40/40 - 0s - loss: 0.0032 - 380ms/epoch - 10ms/step\n",
      "Epoch 395/1000\n",
      "40/40 - 0s - loss: 0.0028 - 335ms/epoch - 8ms/step\n",
      "Epoch 396/1000\n",
      "40/40 - 0s - loss: 0.0067 - 301ms/epoch - 8ms/step\n",
      "Epoch 397/1000\n",
      "40/40 - 0s - loss: 0.0118 - 401ms/epoch - 10ms/step\n",
      "Epoch 398/1000\n",
      "40/40 - 0s - loss: 0.0128 - 415ms/epoch - 10ms/step\n",
      "Epoch 399/1000\n",
      "40/40 - 0s - loss: 0.0098 - 328ms/epoch - 8ms/step\n",
      "Epoch 400/1000\n",
      "40/40 - 0s - loss: 0.0040 - 420ms/epoch - 11ms/step\n",
      "Epoch 401/1000\n",
      "40/40 - 0s - loss: 0.0033 - 439ms/epoch - 11ms/step\n",
      "Epoch 402/1000\n",
      "40/40 - 0s - loss: 0.0027 - 441ms/epoch - 11ms/step\n",
      "Epoch 403/1000\n",
      "40/40 - 0s - loss: 0.0032 - 338ms/epoch - 8ms/step\n",
      "Epoch 404/1000\n",
      "40/40 - 0s - loss: 0.0042 - 370ms/epoch - 9ms/step\n",
      "Epoch 405/1000\n",
      "40/40 - 0s - loss: 0.0030 - 358ms/epoch - 9ms/step\n",
      "Epoch 406/1000\n",
      "40/40 - 0s - loss: 0.0050 - 361ms/epoch - 9ms/step\n",
      "Epoch 407/1000\n",
      "40/40 - 0s - loss: 0.0117 - 336ms/epoch - 8ms/step\n",
      "Epoch 408/1000\n",
      "40/40 - 0s - loss: 0.0107 - 349ms/epoch - 9ms/step\n",
      "Epoch 409/1000\n",
      "40/40 - 0s - loss: 0.0095 - 374ms/epoch - 9ms/step\n",
      "Epoch 410/1000\n",
      "40/40 - 0s - loss: 0.0038 - 359ms/epoch - 9ms/step\n",
      "Epoch 411/1000\n",
      "40/40 - 0s - loss: 0.0045 - 429ms/epoch - 11ms/step\n",
      "Epoch 412/1000\n",
      "40/40 - 0s - loss: 0.0045 - 346ms/epoch - 9ms/step\n",
      "Epoch 413/1000\n",
      "40/40 - 0s - loss: 0.0032 - 333ms/epoch - 8ms/step\n",
      "Epoch 414/1000\n",
      "40/40 - 0s - loss: 0.0030 - 336ms/epoch - 8ms/step\n",
      "Epoch 415/1000\n",
      "40/40 - 0s - loss: 0.0025 - 351ms/epoch - 9ms/step\n",
      "Epoch 416/1000\n",
      "40/40 - 0s - loss: 0.0028 - 345ms/epoch - 9ms/step\n",
      "Epoch 417/1000\n",
      "40/40 - 0s - loss: 0.0047 - 372ms/epoch - 9ms/step\n",
      "Epoch 418/1000\n",
      "40/40 - 0s - loss: 0.0054 - 415ms/epoch - 10ms/step\n",
      "Epoch 419/1000\n",
      "40/40 - 0s - loss: 0.0037 - 361ms/epoch - 9ms/step\n",
      "Epoch 420/1000\n",
      "40/40 - 0s - loss: 0.0025 - 397ms/epoch - 10ms/step\n",
      "Epoch 421/1000\n",
      "40/40 - 0s - loss: 0.0029 - 393ms/epoch - 10ms/step\n",
      "Epoch 422/1000\n",
      "40/40 - 0s - loss: 0.0033 - 345ms/epoch - 9ms/step\n",
      "Epoch 423/1000\n",
      "40/40 - 0s - loss: 0.0025 - 346ms/epoch - 9ms/step\n",
      "Epoch 424/1000\n",
      "40/40 - 0s - loss: 0.0022 - 323ms/epoch - 8ms/step\n",
      "Epoch 425/1000\n",
      "40/40 - 0s - loss: 0.0023 - 346ms/epoch - 9ms/step\n",
      "Epoch 426/1000\n",
      "40/40 - 0s - loss: 0.0022 - 329ms/epoch - 8ms/step\n",
      "Epoch 427/1000\n",
      "40/40 - 0s - loss: 0.0019 - 342ms/epoch - 9ms/step\n",
      "Epoch 428/1000\n",
      "40/40 - 0s - loss: 0.0018 - 378ms/epoch - 9ms/step\n",
      "Epoch 429/1000\n",
      "40/40 - 0s - loss: 0.0022 - 481ms/epoch - 12ms/step\n",
      "Epoch 430/1000\n",
      "40/40 - 0s - loss: 0.0026 - 453ms/epoch - 11ms/step\n",
      "Epoch 431/1000\n",
      "40/40 - 0s - loss: 0.0020 - 445ms/epoch - 11ms/step\n",
      "Epoch 432/1000\n",
      "40/40 - 0s - loss: 0.0026 - 416ms/epoch - 10ms/step\n",
      "Epoch 433/1000\n",
      "40/40 - 0s - loss: 0.0021 - 451ms/epoch - 11ms/step\n",
      "Epoch 434/1000\n",
      "40/40 - 0s - loss: 0.0025 - 472ms/epoch - 12ms/step\n",
      "Epoch 435/1000\n",
      "40/40 - 0s - loss: 0.0020 - 386ms/epoch - 10ms/step\n",
      "Epoch 436/1000\n",
      "40/40 - 0s - loss: 0.0021 - 398ms/epoch - 10ms/step\n",
      "Epoch 437/1000\n",
      "40/40 - 0s - loss: 0.0032 - 363ms/epoch - 9ms/step\n",
      "Epoch 438/1000\n",
      "40/40 - 0s - loss: 0.0039 - 404ms/epoch - 10ms/step\n",
      "Epoch 439/1000\n",
      "40/40 - 0s - loss: 0.0027 - 401ms/epoch - 10ms/step\n",
      "Epoch 440/1000\n",
      "40/40 - 0s - loss: 0.0025 - 375ms/epoch - 9ms/step\n",
      "Epoch 441/1000\n",
      "40/40 - 0s - loss: 0.0021 - 335ms/epoch - 8ms/step\n",
      "Epoch 442/1000\n",
      "40/40 - 0s - loss: 0.0025 - 300ms/epoch - 7ms/step\n",
      "Epoch 443/1000\n",
      "40/40 - 0s - loss: 0.0027 - 307ms/epoch - 8ms/step\n",
      "Epoch 444/1000\n",
      "40/40 - 0s - loss: 0.0017 - 336ms/epoch - 8ms/step\n",
      "Epoch 445/1000\n",
      "40/40 - 0s - loss: 0.0026 - 302ms/epoch - 8ms/step\n",
      "Epoch 446/1000\n",
      "40/40 - 0s - loss: 0.0021 - 310ms/epoch - 8ms/step\n",
      "Epoch 447/1000\n",
      "40/40 - 0s - loss: 0.0032 - 314ms/epoch - 8ms/step\n",
      "Epoch 448/1000\n",
      "40/40 - 0s - loss: 0.0248 - 299ms/epoch - 7ms/step\n",
      "Epoch 449/1000\n",
      "40/40 - 0s - loss: 0.0117 - 315ms/epoch - 8ms/step\n",
      "Epoch 450/1000\n",
      "40/40 - 0s - loss: 0.0062 - 382ms/epoch - 10ms/step\n",
      "Epoch 451/1000\n",
      "40/40 - 0s - loss: 0.0038 - 378ms/epoch - 9ms/step\n",
      "Epoch 452/1000\n",
      "40/40 - 0s - loss: 0.0165 - 380ms/epoch - 10ms/step\n",
      "Epoch 453/1000\n",
      "40/40 - 0s - loss: 0.0233 - 424ms/epoch - 11ms/step\n",
      "Epoch 454/1000\n",
      "40/40 - 0s - loss: 0.0170 - 351ms/epoch - 9ms/step\n",
      "Epoch 455/1000\n",
      "40/40 - 0s - loss: 0.0075 - 331ms/epoch - 8ms/step\n",
      "Epoch 456/1000\n",
      "40/40 - 0s - loss: 0.0052 - 334ms/epoch - 8ms/step\n",
      "Epoch 457/1000\n",
      "40/40 - 0s - loss: 0.0134 - 363ms/epoch - 9ms/step\n",
      "Epoch 458/1000\n",
      "40/40 - 0s - loss: 0.0043 - 362ms/epoch - 9ms/step\n",
      "Epoch 459/1000\n",
      "40/40 - 0s - loss: 0.0029 - 346ms/epoch - 9ms/step\n",
      "Epoch 460/1000\n",
      "40/40 - 0s - loss: 0.0025 - 382ms/epoch - 10ms/step\n",
      "Epoch 461/1000\n",
      "40/40 - 0s - loss: 0.0038 - 360ms/epoch - 9ms/step\n",
      "Epoch 462/1000\n",
      "40/40 - 0s - loss: 0.0028 - 379ms/epoch - 9ms/step\n",
      "Epoch 463/1000\n",
      "40/40 - 0s - loss: 0.0024 - 380ms/epoch - 9ms/step\n",
      "Epoch 464/1000\n",
      "40/40 - 0s - loss: 0.0021 - 341ms/epoch - 9ms/step\n",
      "Epoch 465/1000\n",
      "40/40 - 0s - loss: 0.0022 - 348ms/epoch - 9ms/step\n",
      "Epoch 466/1000\n",
      "40/40 - 0s - loss: 0.0019 - 385ms/epoch - 10ms/step\n",
      "Epoch 467/1000\n",
      "40/40 - 0s - loss: 0.0016 - 317ms/epoch - 8ms/step\n",
      "Epoch 468/1000\n",
      "40/40 - 0s - loss: 0.0016 - 300ms/epoch - 7ms/step\n",
      "Epoch 469/1000\n",
      "40/40 - 0s - loss: 0.0073 - 316ms/epoch - 8ms/step\n",
      "Epoch 470/1000\n",
      "40/40 - 0s - loss: 0.0096 - 309ms/epoch - 8ms/step\n",
      "Epoch 471/1000\n",
      "40/40 - 0s - loss: 0.0038 - 334ms/epoch - 8ms/step\n",
      "Epoch 472/1000\n",
      "40/40 - 0s - loss: 0.0071 - 377ms/epoch - 9ms/step\n",
      "Epoch 473/1000\n",
      "40/40 - 0s - loss: 0.0226 - 443ms/epoch - 11ms/step\n",
      "Epoch 474/1000\n",
      "40/40 - 0s - loss: 0.0069 - 418ms/epoch - 10ms/step\n",
      "Epoch 475/1000\n",
      "40/40 - 0s - loss: 0.0039 - 442ms/epoch - 11ms/step\n",
      "Epoch 476/1000\n",
      "40/40 - 0s - loss: 0.0027 - 431ms/epoch - 11ms/step\n",
      "Epoch 477/1000\n",
      "40/40 - 0s - loss: 0.0027 - 458ms/epoch - 11ms/step\n",
      "Epoch 478/1000\n",
      "40/40 - 0s - loss: 0.0020 - 434ms/epoch - 11ms/step\n",
      "Epoch 479/1000\n",
      "40/40 - 0s - loss: 0.0019 - 410ms/epoch - 10ms/step\n",
      "Epoch 480/1000\n",
      "40/40 - 0s - loss: 0.0023 - 392ms/epoch - 10ms/step\n",
      "Epoch 481/1000\n",
      "40/40 - 0s - loss: 0.0024 - 345ms/epoch - 9ms/step\n",
      "Epoch 482/1000\n",
      "40/40 - 0s - loss: 0.0017 - 331ms/epoch - 8ms/step\n",
      "Epoch 483/1000\n",
      "40/40 - 0s - loss: 0.0020 - 303ms/epoch - 8ms/step\n",
      "Epoch 484/1000\n",
      "40/40 - 0s - loss: 0.0019 - 305ms/epoch - 8ms/step\n",
      "Epoch 485/1000\n",
      "40/40 - 0s - loss: 0.0018 - 308ms/epoch - 8ms/step\n",
      "Epoch 486/1000\n",
      "40/40 - 0s - loss: 0.0017 - 311ms/epoch - 8ms/step\n",
      "Epoch 487/1000\n",
      "40/40 - 0s - loss: 0.0022 - 316ms/epoch - 8ms/step\n",
      "Epoch 488/1000\n",
      "40/40 - 0s - loss: 0.0016 - 326ms/epoch - 8ms/step\n",
      "Epoch 489/1000\n",
      "40/40 - 0s - loss: 0.0025 - 306ms/epoch - 8ms/step\n",
      "Epoch 490/1000\n",
      "40/40 - 0s - loss: 0.0019 - 319ms/epoch - 8ms/step\n",
      "Epoch 491/1000\n",
      "40/40 - 0s - loss: 0.0022 - 346ms/epoch - 9ms/step\n",
      "Epoch 492/1000\n",
      "40/40 - 0s - loss: 0.0023 - 402ms/epoch - 10ms/step\n",
      "Epoch 493/1000\n",
      "40/40 - 0s - loss: 0.0033 - 400ms/epoch - 10ms/step\n",
      "Epoch 494/1000\n",
      "40/40 - 0s - loss: 0.0021 - 359ms/epoch - 9ms/step\n",
      "Epoch 495/1000\n",
      "40/40 - 0s - loss: 0.0019 - 336ms/epoch - 8ms/step\n",
      "Epoch 496/1000\n",
      "40/40 - 0s - loss: 0.0021 - 380ms/epoch - 10ms/step\n",
      "Epoch 497/1000\n",
      "40/40 - 0s - loss: 0.0021 - 366ms/epoch - 9ms/step\n",
      "Epoch 498/1000\n",
      "40/40 - 0s - loss: 0.0019 - 314ms/epoch - 8ms/step\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 499/1000\n",
      "40/40 - 0s - loss: 0.0019 - 355ms/epoch - 9ms/step\n",
      "Epoch 500/1000\n",
      "40/40 - 0s - loss: 0.0016 - 439ms/epoch - 11ms/step\n",
      "Epoch 501/1000\n",
      "40/40 - 0s - loss: 0.0014 - 385ms/epoch - 10ms/step\n",
      "Epoch 502/1000\n",
      "40/40 - 0s - loss: 0.0015 - 391ms/epoch - 10ms/step\n",
      "Epoch 503/1000\n",
      "40/40 - 0s - loss: 0.0013 - 384ms/epoch - 10ms/step\n",
      "Epoch 504/1000\n",
      "40/40 - 0s - loss: 0.0012 - 415ms/epoch - 10ms/step\n",
      "Epoch 505/1000\n",
      "40/40 - 0s - loss: 0.0017 - 403ms/epoch - 10ms/step\n",
      "Epoch 506/1000\n",
      "40/40 - 0s - loss: 0.0024 - 367ms/epoch - 9ms/step\n",
      "Epoch 507/1000\n",
      "40/40 - 0s - loss: 0.0015 - 314ms/epoch - 8ms/step\n",
      "Epoch 508/1000\n",
      "40/40 - 0s - loss: 0.0064 - 297ms/epoch - 7ms/step\n",
      "Epoch 509/1000\n",
      "40/40 - 0s - loss: 0.0097 - 319ms/epoch - 8ms/step\n",
      "Epoch 510/1000\n",
      "40/40 - 0s - loss: 0.0196 - 348ms/epoch - 9ms/step\n",
      "Epoch 511/1000\n",
      "40/40 - 0s - loss: 0.0205 - 371ms/epoch - 9ms/step\n",
      "Epoch 512/1000\n",
      "40/40 - 0s - loss: 0.0114 - 346ms/epoch - 9ms/step\n",
      "Epoch 513/1000\n",
      "40/40 - 0s - loss: 0.0053 - 325ms/epoch - 8ms/step\n",
      "Epoch 514/1000\n",
      "40/40 - 0s - loss: 0.0033 - 318ms/epoch - 8ms/step\n",
      "Epoch 515/1000\n",
      "40/40 - 0s - loss: 0.0050 - 330ms/epoch - 8ms/step\n",
      "Epoch 516/1000\n",
      "40/40 - 0s - loss: 0.0061 - 387ms/epoch - 10ms/step\n",
      "Epoch 517/1000\n",
      "40/40 - 0s - loss: 0.0196 - 361ms/epoch - 9ms/step\n",
      "Epoch 518/1000\n",
      "40/40 - 0s - loss: 0.0213 - 406ms/epoch - 10ms/step\n",
      "Epoch 519/1000\n",
      "40/40 - 0s - loss: 0.0162 - 473ms/epoch - 12ms/step\n",
      "Epoch 520/1000\n",
      "40/40 - 0s - loss: 0.0136 - 434ms/epoch - 11ms/step\n",
      "Epoch 521/1000\n",
      "40/40 - 0s - loss: 0.0139 - 369ms/epoch - 9ms/step\n",
      "Epoch 522/1000\n",
      "40/40 - 0s - loss: 0.0130 - 356ms/epoch - 9ms/step\n",
      "Epoch 523/1000\n",
      "40/40 - 0s - loss: 0.0126 - 309ms/epoch - 8ms/step\n",
      "Epoch 524/1000\n",
      "40/40 - 0s - loss: 0.0122 - 309ms/epoch - 8ms/step\n",
      "Epoch 525/1000\n",
      "40/40 - 0s - loss: 0.0116 - 321ms/epoch - 8ms/step\n",
      "Epoch 526/1000\n",
      "40/40 - 0s - loss: 0.0191 - 311ms/epoch - 8ms/step\n",
      "Epoch 527/1000\n",
      "40/40 - 0s - loss: 0.0234 - 310ms/epoch - 8ms/step\n",
      "Epoch 528/1000\n",
      "40/40 - 0s - loss: 0.0415 - 303ms/epoch - 8ms/step\n",
      "Epoch 529/1000\n",
      "40/40 - 0s - loss: 0.0272 - 310ms/epoch - 8ms/step\n",
      "Epoch 530/1000\n",
      "40/40 - 0s - loss: 0.0187 - 321ms/epoch - 8ms/step\n",
      "Epoch 531/1000\n",
      "40/40 - 0s - loss: 0.0116 - 382ms/epoch - 10ms/step\n",
      "Epoch 532/1000\n",
      "40/40 - 0s - loss: 0.0069 - 354ms/epoch - 9ms/step\n",
      "Epoch 533/1000\n",
      "40/40 - 0s - loss: 0.0072 - 342ms/epoch - 9ms/step\n",
      "Epoch 534/1000\n",
      "40/40 - 0s - loss: 0.0063 - 367ms/epoch - 9ms/step\n",
      "Epoch 535/1000\n",
      "40/40 - 0s - loss: 0.0066 - 307ms/epoch - 8ms/step\n",
      "Epoch 536/1000\n",
      "40/40 - 0s - loss: 0.0038 - 350ms/epoch - 9ms/step\n",
      "Epoch 537/1000\n",
      "40/40 - 0s - loss: 0.0039 - 309ms/epoch - 8ms/step\n",
      "Epoch 538/1000\n",
      "40/40 - 0s - loss: 0.0031 - 313ms/epoch - 8ms/step\n",
      "Epoch 539/1000\n",
      "40/40 - 0s - loss: 0.0026 - 333ms/epoch - 8ms/step\n",
      "Epoch 540/1000\n",
      "40/40 - 0s - loss: 0.0027 - 331ms/epoch - 8ms/step\n",
      "Epoch 541/1000\n",
      "40/40 - 0s - loss: 0.0026 - 332ms/epoch - 8ms/step\n",
      "Epoch 542/1000\n",
      "40/40 - 0s - loss: 0.0071 - 305ms/epoch - 8ms/step\n",
      "Epoch 543/1000\n",
      "40/40 - 0s - loss: 0.0027 - 334ms/epoch - 8ms/step\n",
      "Epoch 544/1000\n",
      "40/40 - 0s - loss: 0.0018 - 342ms/epoch - 9ms/step\n",
      "Epoch 545/1000\n",
      "40/40 - 0s - loss: 0.0017 - 330ms/epoch - 8ms/step\n",
      "Epoch 546/1000\n",
      "40/40 - 0s - loss: 0.0016 - 336ms/epoch - 8ms/step\n",
      "Epoch 547/1000\n",
      "40/40 - 0s - loss: 0.0018 - 330ms/epoch - 8ms/step\n",
      "Epoch 548/1000\n",
      "40/40 - 0s - loss: 0.0017 - 339ms/epoch - 8ms/step\n",
      "Epoch 549/1000\n",
      "40/40 - 0s - loss: 0.0015 - 330ms/epoch - 8ms/step\n",
      "Epoch 550/1000\n",
      "40/40 - 0s - loss: 0.0023 - 359ms/epoch - 9ms/step\n",
      "Epoch 551/1000\n",
      "40/40 - 0s - loss: 0.0015 - 314ms/epoch - 8ms/step\n",
      "Epoch 552/1000\n",
      "40/40 - 0s - loss: 0.0015 - 326ms/epoch - 8ms/step\n",
      "Epoch 553/1000\n",
      "40/40 - 0s - loss: 0.0019 - 340ms/epoch - 8ms/step\n",
      "Epoch 554/1000\n",
      "40/40 - 0s - loss: 0.0022 - 344ms/epoch - 9ms/step\n",
      "Epoch 555/1000\n",
      "40/40 - 0s - loss: 0.0017 - 337ms/epoch - 8ms/step\n",
      "Epoch 556/1000\n",
      "40/40 - 0s - loss: 0.0014 - 347ms/epoch - 9ms/step\n",
      "Epoch 557/1000\n",
      "40/40 - 0s - loss: 0.0015 - 339ms/epoch - 8ms/step\n",
      "Epoch 558/1000\n",
      "40/40 - 0s - loss: 0.0013 - 316ms/epoch - 8ms/step\n",
      "Epoch 559/1000\n",
      "40/40 - 0s - loss: 0.0014 - 342ms/epoch - 9ms/step\n",
      "Epoch 560/1000\n",
      "40/40 - 0s - loss: 0.0016 - 328ms/epoch - 8ms/step\n",
      "Epoch 561/1000\n",
      "40/40 - 0s - loss: 0.0018 - 316ms/epoch - 8ms/step\n",
      "Epoch 562/1000\n",
      "40/40 - 0s - loss: 0.0012 - 303ms/epoch - 8ms/step\n",
      "Epoch 563/1000\n",
      "40/40 - 0s - loss: 0.0011 - 333ms/epoch - 8ms/step\n",
      "Epoch 564/1000\n",
      "40/40 - 0s - loss: 0.0014 - 380ms/epoch - 9ms/step\n",
      "Epoch 565/1000\n",
      "40/40 - 0s - loss: 0.0015 - 361ms/epoch - 9ms/step\n",
      "Epoch 566/1000\n",
      "40/40 - 0s - loss: 0.0016 - 362ms/epoch - 9ms/step\n",
      "Epoch 567/1000\n",
      "40/40 - 0s - loss: 0.0011 - 370ms/epoch - 9ms/step\n",
      "Epoch 568/1000\n",
      "40/40 - 0s - loss: 0.0012 - 372ms/epoch - 9ms/step\n",
      "Epoch 569/1000\n",
      "40/40 - 0s - loss: 0.0043 - 340ms/epoch - 8ms/step\n",
      "Epoch 570/1000\n",
      "40/40 - 0s - loss: 0.0029 - 355ms/epoch - 9ms/step\n",
      "Epoch 571/1000\n",
      "40/40 - 0s - loss: 0.0029 - 342ms/epoch - 9ms/step\n",
      "Epoch 572/1000\n",
      "40/40 - 0s - loss: 0.0022 - 372ms/epoch - 9ms/step\n",
      "Epoch 573/1000\n",
      "40/40 - 0s - loss: 0.0017 - 307ms/epoch - 8ms/step\n",
      "Epoch 574/1000\n",
      "40/40 - 0s - loss: 0.0016 - 320ms/epoch - 8ms/step\n",
      "Epoch 575/1000\n",
      "40/40 - 0s - loss: 0.0014 - 306ms/epoch - 8ms/step\n",
      "Epoch 576/1000\n",
      "40/40 - 0s - loss: 0.0062 - 340ms/epoch - 9ms/step\n",
      "Epoch 577/1000\n",
      "40/40 - 0s - loss: 0.0051 - 323ms/epoch - 8ms/step\n",
      "Epoch 578/1000\n",
      "40/40 - 0s - loss: 0.0109 - 386ms/epoch - 10ms/step\n",
      "Epoch 579/1000\n",
      "40/40 - 0s - loss: 0.0089 - 365ms/epoch - 9ms/step\n",
      "Epoch 580/1000\n",
      "40/40 - 0s - loss: 0.0155 - 322ms/epoch - 8ms/step\n",
      "Epoch 581/1000\n",
      "40/40 - 0s - loss: 0.0063 - 324ms/epoch - 8ms/step\n",
      "Epoch 582/1000\n",
      "40/40 - 0s - loss: 0.0042 - 321ms/epoch - 8ms/step\n",
      "Epoch 583/1000\n",
      "40/40 - 0s - loss: 0.0044 - 367ms/epoch - 9ms/step\n",
      "Epoch 584/1000\n",
      "40/40 - 0s - loss: 0.0041 - 306ms/epoch - 8ms/step\n",
      "Epoch 585/1000\n",
      "40/40 - 0s - loss: 0.0020 - 378ms/epoch - 9ms/step\n",
      "Epoch 586/1000\n",
      "40/40 - 0s - loss: 0.0016 - 316ms/epoch - 8ms/step\n",
      "Epoch 587/1000\n",
      "40/40 - 0s - loss: 0.0015 - 335ms/epoch - 8ms/step\n",
      "Epoch 588/1000\n",
      "40/40 - 0s - loss: 0.0013 - 330ms/epoch - 8ms/step\n",
      "Epoch 589/1000\n",
      "40/40 - 0s - loss: 0.0013 - 328ms/epoch - 8ms/step\n",
      "Epoch 590/1000\n",
      "40/40 - 0s - loss: 0.0013 - 308ms/epoch - 8ms/step\n",
      "Epoch 591/1000\n",
      "40/40 - 0s - loss: 0.0011 - 349ms/epoch - 9ms/step\n",
      "Epoch 592/1000\n",
      "40/40 - 0s - loss: 0.0011 - 343ms/epoch - 9ms/step\n",
      "Epoch 593/1000\n",
      "40/40 - 0s - loss: 0.0012 - 306ms/epoch - 8ms/step\n",
      "Epoch 594/1000\n",
      "40/40 - 0s - loss: 0.0018 - 310ms/epoch - 8ms/step\n",
      "Epoch 595/1000\n",
      "40/40 - 0s - loss: 0.0019 - 318ms/epoch - 8ms/step\n",
      "Epoch 596/1000\n",
      "40/40 - 0s - loss: 0.0018 - 336ms/epoch - 8ms/step\n",
      "Epoch 597/1000\n",
      "40/40 - 0s - loss: 0.0018 - 328ms/epoch - 8ms/step\n",
      "Epoch 598/1000\n",
      "40/40 - 0s - loss: 0.0016 - 334ms/epoch - 8ms/step\n",
      "Epoch 599/1000\n",
      "40/40 - 0s - loss: 0.0015 - 381ms/epoch - 10ms/step\n",
      "Epoch 600/1000\n",
      "40/40 - 0s - loss: 0.0014 - 324ms/epoch - 8ms/step\n",
      "Epoch 601/1000\n",
      "40/40 - 0s - loss: 0.0015 - 321ms/epoch - 8ms/step\n",
      "Epoch 602/1000\n",
      "40/40 - 0s - loss: 0.0013 - 321ms/epoch - 8ms/step\n",
      "Epoch 603/1000\n",
      "40/40 - 0s - loss: 0.0013 - 327ms/epoch - 8ms/step\n",
      "Epoch 604/1000\n",
      "40/40 - 0s - loss: 0.0014 - 358ms/epoch - 9ms/step\n",
      "Epoch 605/1000\n",
      "40/40 - 0s - loss: 0.0020 - 397ms/epoch - 10ms/step\n",
      "Epoch 606/1000\n",
      "40/40 - 0s - loss: 0.0014 - 344ms/epoch - 9ms/step\n",
      "Epoch 607/1000\n",
      "40/40 - 0s - loss: 0.0012 - 342ms/epoch - 9ms/step\n",
      "Epoch 608/1000\n",
      "40/40 - 0s - loss: 0.0016 - 348ms/epoch - 9ms/step\n",
      "Epoch 609/1000\n",
      "40/40 - 0s - loss: 0.0015 - 343ms/epoch - 9ms/step\n",
      "Epoch 610/1000\n",
      "40/40 - 0s - loss: 0.0014 - 352ms/epoch - 9ms/step\n",
      "Epoch 611/1000\n",
      "40/40 - 0s - loss: 0.0011 - 372ms/epoch - 9ms/step\n",
      "Epoch 612/1000\n",
      "40/40 - 0s - loss: 0.0016 - 365ms/epoch - 9ms/step\n",
      "Epoch 613/1000\n",
      "40/40 - 0s - loss: 0.0012 - 380ms/epoch - 9ms/step\n",
      "Epoch 614/1000\n",
      "40/40 - 0s - loss: 0.0010 - 378ms/epoch - 9ms/step\n",
      "Epoch 615/1000\n",
      "40/40 - 0s - loss: 0.0010 - 380ms/epoch - 9ms/step\n",
      "Epoch 616/1000\n",
      "40/40 - 0s - loss: 0.0016 - 384ms/epoch - 10ms/step\n",
      "Epoch 617/1000\n",
      "40/40 - 0s - loss: 0.0012 - 340ms/epoch - 8ms/step\n",
      "Epoch 618/1000\n",
      "40/40 - 0s - loss: 0.0015 - 324ms/epoch - 8ms/step\n",
      "Epoch 619/1000\n",
      "40/40 - 0s - loss: 0.0016 - 342ms/epoch - 9ms/step\n",
      "Epoch 620/1000\n",
      "40/40 - 0s - loss: 0.0030 - 345ms/epoch - 9ms/step\n",
      "Epoch 621/1000\n",
      "40/40 - 0s - loss: 0.0091 - 323ms/epoch - 8ms/step\n",
      "Epoch 622/1000\n",
      "40/40 - 0s - loss: 0.0295 - 359ms/epoch - 9ms/step\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 623/1000\n",
      "40/40 - 0s - loss: 0.0144 - 358ms/epoch - 9ms/step\n",
      "Epoch 624/1000\n",
      "40/40 - 0s - loss: 0.0192 - 360ms/epoch - 9ms/step\n",
      "Epoch 625/1000\n",
      "40/40 - 0s - loss: 0.0066 - 334ms/epoch - 8ms/step\n",
      "Epoch 626/1000\n",
      "40/40 - 0s - loss: 0.0033 - 358ms/epoch - 9ms/step\n",
      "Epoch 627/1000\n",
      "40/40 - 0s - loss: 0.0018 - 331ms/epoch - 8ms/step\n",
      "Epoch 628/1000\n",
      "40/40 - 0s - loss: 0.0017 - 342ms/epoch - 9ms/step\n",
      "Epoch 629/1000\n",
      "40/40 - 0s - loss: 0.0016 - 349ms/epoch - 9ms/step\n",
      "Epoch 630/1000\n",
      "40/40 - 0s - loss: 0.0016 - 352ms/epoch - 9ms/step\n",
      "Epoch 631/1000\n",
      "40/40 - 0s - loss: 0.0013 - 339ms/epoch - 8ms/step\n",
      "Epoch 632/1000\n",
      "40/40 - 0s - loss: 0.0013 - 369ms/epoch - 9ms/step\n",
      "Epoch 633/1000\n",
      "40/40 - 0s - loss: 0.0011 - 315ms/epoch - 8ms/step\n",
      "Epoch 634/1000\n",
      "40/40 - 0s - loss: 0.0015 - 334ms/epoch - 8ms/step\n",
      "Epoch 635/1000\n",
      "40/40 - 0s - loss: 0.0012 - 335ms/epoch - 8ms/step\n",
      "Epoch 636/1000\n",
      "40/40 - 0s - loss: 0.0011 - 338ms/epoch - 8ms/step\n",
      "Epoch 637/1000\n",
      "40/40 - 0s - loss: 0.0012 - 350ms/epoch - 9ms/step\n",
      "Epoch 638/1000\n",
      "40/40 - 0s - loss: 0.0010 - 325ms/epoch - 8ms/step\n",
      "Epoch 639/1000\n",
      "40/40 - 0s - loss: 0.0010 - 339ms/epoch - 8ms/step\n",
      "Epoch 640/1000\n",
      "40/40 - 0s - loss: 0.0010 - 340ms/epoch - 9ms/step\n",
      "Epoch 641/1000\n",
      "40/40 - 0s - loss: 9.7167e-04 - 370ms/epoch - 9ms/step\n",
      "Epoch 642/1000\n",
      "40/40 - 0s - loss: 0.0011 - 327ms/epoch - 8ms/step\n",
      "Epoch 643/1000\n",
      "40/40 - 0s - loss: 9.5330e-04 - 336ms/epoch - 8ms/step\n",
      "Epoch 644/1000\n",
      "40/40 - 0s - loss: 9.8007e-04 - 315ms/epoch - 8ms/step\n",
      "Epoch 645/1000\n",
      "40/40 - 0s - loss: 9.1952e-04 - 340ms/epoch - 8ms/step\n",
      "Epoch 646/1000\n",
      "40/40 - 0s - loss: 0.0010 - 320ms/epoch - 8ms/step\n",
      "Epoch 647/1000\n",
      "40/40 - 0s - loss: 0.0012 - 310ms/epoch - 8ms/step\n",
      "Epoch 648/1000\n",
      "40/40 - 0s - loss: 8.7386e-04 - 337ms/epoch - 8ms/step\n",
      "Epoch 649/1000\n",
      "40/40 - 0s - loss: 0.0012 - 359ms/epoch - 9ms/step\n",
      "Epoch 650/1000\n",
      "40/40 - 0s - loss: 0.0010 - 379ms/epoch - 9ms/step\n",
      "Epoch 651/1000\n",
      "40/40 - 0s - loss: 8.8580e-04 - 343ms/epoch - 9ms/step\n",
      "Epoch 652/1000\n",
      "40/40 - 0s - loss: 0.0010 - 345ms/epoch - 9ms/step\n",
      "Epoch 653/1000\n",
      "40/40 - 0s - loss: 9.1516e-04 - 331ms/epoch - 8ms/step\n",
      "Epoch 654/1000\n",
      "40/40 - 0s - loss: 0.0010 - 331ms/epoch - 8ms/step\n",
      "Epoch 655/1000\n",
      "40/40 - 0s - loss: 0.0020 - 370ms/epoch - 9ms/step\n",
      "Epoch 656/1000\n",
      "40/40 - 0s - loss: 0.0012 - 370ms/epoch - 9ms/step\n",
      "Epoch 657/1000\n",
      "40/40 - 0s - loss: 9.3929e-04 - 379ms/epoch - 9ms/step\n",
      "Epoch 658/1000\n",
      "40/40 - 0s - loss: 0.0012 - 351ms/epoch - 9ms/step\n",
      "Epoch 659/1000\n",
      "40/40 - 0s - loss: 9.4080e-04 - 395ms/epoch - 10ms/step\n",
      "Epoch 660/1000\n",
      "40/40 - 0s - loss: 0.0015 - 392ms/epoch - 10ms/step\n",
      "Epoch 661/1000\n",
      "40/40 - 0s - loss: 0.0011 - 363ms/epoch - 9ms/step\n",
      "Epoch 662/1000\n",
      "40/40 - 0s - loss: 9.2237e-04 - 358ms/epoch - 9ms/step\n",
      "Epoch 663/1000\n",
      "40/40 - 0s - loss: 0.0010 - 324ms/epoch - 8ms/step\n",
      "Epoch 664/1000\n",
      "40/40 - 0s - loss: 0.0015 - 354ms/epoch - 9ms/step\n",
      "Epoch 665/1000\n",
      "40/40 - 0s - loss: 0.0020 - 354ms/epoch - 9ms/step\n",
      "Epoch 666/1000\n",
      "40/40 - 0s - loss: 0.0011 - 349ms/epoch - 9ms/step\n",
      "Epoch 667/1000\n",
      "40/40 - 0s - loss: 9.6104e-04 - 354ms/epoch - 9ms/step\n",
      "Epoch 668/1000\n",
      "40/40 - 0s - loss: 8.9898e-04 - 339ms/epoch - 8ms/step\n",
      "Epoch 669/1000\n",
      "40/40 - 0s - loss: 9.1175e-04 - 338ms/epoch - 8ms/step\n",
      "Epoch 670/1000\n",
      "40/40 - 0s - loss: 0.0011 - 317ms/epoch - 8ms/step\n",
      "Epoch 671/1000\n",
      "40/40 - 0s - loss: 8.9793e-04 - 358ms/epoch - 9ms/step\n",
      "Epoch 672/1000\n",
      "40/40 - 0s - loss: 8.3069e-04 - 352ms/epoch - 9ms/step\n",
      "Epoch 673/1000\n",
      "40/40 - 0s - loss: 0.0013 - 329ms/epoch - 8ms/step\n",
      "Epoch 674/1000\n",
      "40/40 - 0s - loss: 0.0012 - 309ms/epoch - 8ms/step\n",
      "Epoch 675/1000\n",
      "40/40 - 0s - loss: 0.0012 - 319ms/epoch - 8ms/step\n",
      "Epoch 676/1000\n",
      "40/40 - 0s - loss: 9.0966e-04 - 322ms/epoch - 8ms/step\n",
      "Epoch 677/1000\n",
      "40/40 - 0s - loss: 0.0012 - 339ms/epoch - 8ms/step\n",
      "Epoch 678/1000\n",
      "40/40 - 0s - loss: 9.5620e-04 - 312ms/epoch - 8ms/step\n",
      "Epoch 679/1000\n",
      "40/40 - 0s - loss: 8.1352e-04 - 319ms/epoch - 8ms/step\n",
      "Epoch 680/1000\n",
      "40/40 - 0s - loss: 0.0012 - 330ms/epoch - 8ms/step\n",
      "Epoch 681/1000\n",
      "40/40 - 0s - loss: 0.0027 - 341ms/epoch - 9ms/step\n",
      "Epoch 682/1000\n",
      "40/40 - 0s - loss: 0.0013 - 319ms/epoch - 8ms/step\n",
      "Epoch 683/1000\n",
      "40/40 - 0s - loss: 9.4508e-04 - 332ms/epoch - 8ms/step\n",
      "Epoch 684/1000\n",
      "40/40 - 0s - loss: 0.0012 - 343ms/epoch - 9ms/step\n",
      "Epoch 685/1000\n",
      "40/40 - 0s - loss: 0.0013 - 342ms/epoch - 9ms/step\n",
      "Epoch 686/1000\n",
      "40/40 - 0s - loss: 0.0011 - 406ms/epoch - 10ms/step\n",
      "Epoch 687/1000\n",
      "40/40 - 0s - loss: 0.0010 - 369ms/epoch - 9ms/step\n",
      "Epoch 688/1000\n",
      "40/40 - 0s - loss: 8.0191e-04 - 418ms/epoch - 10ms/step\n",
      "Epoch 689/1000\n",
      "40/40 - 0s - loss: 7.8153e-04 - 345ms/epoch - 9ms/step\n",
      "Epoch 690/1000\n",
      "40/40 - 0s - loss: 0.0011 - 326ms/epoch - 8ms/step\n",
      "Epoch 691/1000\n",
      "40/40 - 0s - loss: 0.0013 - 330ms/epoch - 8ms/step\n",
      "Epoch 692/1000\n",
      "40/40 - 0s - loss: 0.0011 - 359ms/epoch - 9ms/step\n",
      "Epoch 693/1000\n",
      "40/40 - 0s - loss: 0.0010 - 354ms/epoch - 9ms/step\n",
      "Epoch 694/1000\n",
      "40/40 - 0s - loss: 0.0015 - 375ms/epoch - 9ms/step\n",
      "Epoch 695/1000\n",
      "40/40 - 0s - loss: 0.0094 - 342ms/epoch - 9ms/step\n",
      "Epoch 696/1000\n",
      "40/40 - 0s - loss: 0.0130 - 343ms/epoch - 9ms/step\n",
      "Epoch 697/1000\n",
      "40/40 - 0s - loss: 0.0128 - 360ms/epoch - 9ms/step\n",
      "Epoch 698/1000\n",
      "40/40 - 0s - loss: 0.0030 - 332ms/epoch - 8ms/step\n",
      "Epoch 699/1000\n",
      "40/40 - 0s - loss: 0.0013 - 382ms/epoch - 10ms/step\n",
      "Epoch 700/1000\n",
      "40/40 - 0s - loss: 9.7714e-04 - 334ms/epoch - 8ms/step\n",
      "Epoch 701/1000\n",
      "40/40 - 0s - loss: 0.0011 - 390ms/epoch - 10ms/step\n",
      "Epoch 702/1000\n",
      "40/40 - 0s - loss: 0.0014 - 418ms/epoch - 10ms/step\n",
      "Epoch 703/1000\n",
      "40/40 - 0s - loss: 0.0012 - 444ms/epoch - 11ms/step\n",
      "Epoch 704/1000\n",
      "40/40 - 0s - loss: 0.0016 - 443ms/epoch - 11ms/step\n",
      "Epoch 705/1000\n",
      "40/40 - 0s - loss: 0.0016 - 434ms/epoch - 11ms/step\n",
      "Epoch 706/1000\n",
      "40/40 - 0s - loss: 0.0012 - 395ms/epoch - 10ms/step\n",
      "Epoch 707/1000\n",
      "40/40 - 0s - loss: 0.0011 - 397ms/epoch - 10ms/step\n",
      "Epoch 708/1000\n",
      "40/40 - 0s - loss: 9.2714e-04 - 355ms/epoch - 9ms/step\n",
      "Epoch 709/1000\n",
      "40/40 - 0s - loss: 0.0010 - 320ms/epoch - 8ms/step\n",
      "Epoch 710/1000\n",
      "40/40 - 0s - loss: 7.9548e-04 - 346ms/epoch - 9ms/step\n",
      "Epoch 711/1000\n",
      "40/40 - 0s - loss: 0.0011 - 368ms/epoch - 9ms/step\n",
      "Epoch 712/1000\n",
      "40/40 - 0s - loss: 0.0012 - 378ms/epoch - 9ms/step\n",
      "Epoch 713/1000\n",
      "40/40 - 0s - loss: 0.0052 - 331ms/epoch - 8ms/step\n",
      "Epoch 714/1000\n",
      "40/40 - 0s - loss: 0.0019 - 376ms/epoch - 9ms/step\n",
      "Epoch 715/1000\n",
      "40/40 - 0s - loss: 0.0016 - 373ms/epoch - 9ms/step\n",
      "Epoch 716/1000\n",
      "40/40 - 0s - loss: 0.0031 - 406ms/epoch - 10ms/step\n",
      "Epoch 717/1000\n",
      "40/40 - 0s - loss: 0.0282 - 355ms/epoch - 9ms/step\n",
      "Epoch 718/1000\n",
      "40/40 - 0s - loss: 0.0335 - 353ms/epoch - 9ms/step\n",
      "Epoch 719/1000\n",
      "40/40 - 0s - loss: 0.0233 - 336ms/epoch - 8ms/step\n",
      "Epoch 720/1000\n",
      "40/40 - 0s - loss: 0.0236 - 341ms/epoch - 9ms/step\n",
      "Epoch 721/1000\n",
      "40/40 - 0s - loss: 0.0145 - 459ms/epoch - 11ms/step\n",
      "Epoch 722/1000\n",
      "40/40 - 0s - loss: 0.0135 - 377ms/epoch - 9ms/step\n",
      "Epoch 723/1000\n",
      "40/40 - 0s - loss: 0.0115 - 344ms/epoch - 9ms/step\n",
      "Epoch 724/1000\n",
      "40/40 - 0s - loss: 0.0116 - 350ms/epoch - 9ms/step\n",
      "Epoch 725/1000\n",
      "40/40 - 0s - loss: 0.0105 - 329ms/epoch - 8ms/step\n",
      "Epoch 726/1000\n",
      "40/40 - 0s - loss: 0.0132 - 346ms/epoch - 9ms/step\n",
      "Epoch 727/1000\n",
      "40/40 - 0s - loss: 0.0136 - 329ms/epoch - 8ms/step\n",
      "Epoch 728/1000\n",
      "40/40 - 0s - loss: 0.0097 - 370ms/epoch - 9ms/step\n",
      "Epoch 729/1000\n",
      "40/40 - 0s - loss: 0.0086 - 341ms/epoch - 9ms/step\n",
      "Epoch 730/1000\n",
      "40/40 - 0s - loss: 0.0089 - 348ms/epoch - 9ms/step\n",
      "Epoch 731/1000\n",
      "40/40 - 0s - loss: 0.0108 - 346ms/epoch - 9ms/step\n",
      "Epoch 732/1000\n",
      "40/40 - 0s - loss: 0.0149 - 347ms/epoch - 9ms/step\n",
      "Epoch 733/1000\n",
      "40/40 - 0s - loss: 0.0108 - 351ms/epoch - 9ms/step\n",
      "Epoch 734/1000\n",
      "40/40 - 0s - loss: 0.0096 - 342ms/epoch - 9ms/step\n",
      "Epoch 735/1000\n",
      "40/40 - 0s - loss: 0.0167 - 328ms/epoch - 8ms/step\n",
      "Epoch 736/1000\n",
      "40/40 - 0s - loss: 0.0081 - 369ms/epoch - 9ms/step\n",
      "Epoch 737/1000\n",
      "40/40 - 0s - loss: 0.0079 - 373ms/epoch - 9ms/step\n",
      "Epoch 738/1000\n",
      "40/40 - 0s - loss: 0.0074 - 333ms/epoch - 8ms/step\n",
      "Epoch 739/1000\n",
      "40/40 - 0s - loss: 0.0081 - 376ms/epoch - 9ms/step\n",
      "Epoch 740/1000\n",
      "40/40 - 0s - loss: 0.0095 - 360ms/epoch - 9ms/step\n",
      "Epoch 741/1000\n",
      "40/40 - 0s - loss: 0.0065 - 351ms/epoch - 9ms/step\n",
      "Epoch 742/1000\n",
      "40/40 - 0s - loss: 0.0066 - 374ms/epoch - 9ms/step\n",
      "Epoch 743/1000\n",
      "40/40 - 0s - loss: 0.0065 - 371ms/epoch - 9ms/step\n",
      "Epoch 744/1000\n",
      "40/40 - 0s - loss: 0.0057 - 348ms/epoch - 9ms/step\n",
      "Epoch 745/1000\n",
      "40/40 - 0s - loss: 0.0057 - 359ms/epoch - 9ms/step\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 746/1000\n",
      "40/40 - 0s - loss: 0.0050 - 408ms/epoch - 10ms/step\n",
      "Epoch 747/1000\n",
      "40/40 - 0s - loss: 0.0045 - 387ms/epoch - 10ms/step\n",
      "Epoch 748/1000\n",
      "40/40 - 0s - loss: 0.0039 - 389ms/epoch - 10ms/step\n",
      "Epoch 749/1000\n",
      "40/40 - 0s - loss: 0.0224 - 413ms/epoch - 10ms/step\n",
      "Epoch 750/1000\n",
      "40/40 - 0s - loss: 0.0134 - 444ms/epoch - 11ms/step\n",
      "Epoch 751/1000\n",
      "40/40 - 0s - loss: 0.0073 - 401ms/epoch - 10ms/step\n",
      "Epoch 752/1000\n",
      "40/40 - 0s - loss: 0.0067 - 341ms/epoch - 9ms/step\n",
      "Epoch 753/1000\n",
      "40/40 - 0s - loss: 0.0063 - 339ms/epoch - 8ms/step\n",
      "Epoch 754/1000\n",
      "40/40 - 0s - loss: 0.0054 - 426ms/epoch - 11ms/step\n",
      "Epoch 755/1000\n",
      "40/40 - 0s - loss: 0.0059 - 369ms/epoch - 9ms/step\n",
      "Epoch 756/1000\n",
      "40/40 - 0s - loss: 0.0047 - 343ms/epoch - 9ms/step\n",
      "Epoch 757/1000\n",
      "40/40 - 0s - loss: 0.0167 - 341ms/epoch - 9ms/step\n",
      "Epoch 758/1000\n",
      "40/40 - 0s - loss: 0.0889 - 326ms/epoch - 8ms/step\n",
      "Epoch 759/1000\n",
      "40/40 - 0s - loss: 0.0708 - 322ms/epoch - 8ms/step\n",
      "Epoch 760/1000\n",
      "40/40 - 0s - loss: 0.0378 - 361ms/epoch - 9ms/step\n",
      "Epoch 761/1000\n",
      "40/40 - 0s - loss: 0.0194 - 461ms/epoch - 12ms/step\n",
      "Epoch 762/1000\n",
      "40/40 - 0s - loss: 0.0451 - 369ms/epoch - 9ms/step\n",
      "Epoch 763/1000\n",
      "40/40 - 0s - loss: 0.0292 - 336ms/epoch - 8ms/step\n",
      "Epoch 764/1000\n",
      "40/40 - 0s - loss: 0.0180 - 314ms/epoch - 8ms/step\n",
      "Epoch 765/1000\n",
      "40/40 - 0s - loss: 0.0066 - 312ms/epoch - 8ms/step\n",
      "Epoch 766/1000\n",
      "40/40 - 0s - loss: 0.0058 - 338ms/epoch - 8ms/step\n",
      "Epoch 767/1000\n",
      "40/40 - 0s - loss: 0.0064 - 342ms/epoch - 9ms/step\n",
      "Epoch 768/1000\n",
      "40/40 - 0s - loss: 0.2974 - 322ms/epoch - 8ms/step\n",
      "Epoch 769/1000\n",
      "40/40 - 0s - loss: 0.1218 - 328ms/epoch - 8ms/step\n",
      "Epoch 770/1000\n",
      "40/40 - 0s - loss: 0.0181 - 352ms/epoch - 9ms/step\n",
      "Epoch 771/1000\n",
      "40/40 - 0s - loss: 0.0107 - 336ms/epoch - 8ms/step\n",
      "Epoch 772/1000\n",
      "40/40 - 0s - loss: 0.0075 - 328ms/epoch - 8ms/step\n",
      "Epoch 773/1000\n",
      "40/40 - 0s - loss: 0.0107 - 322ms/epoch - 8ms/step\n",
      "Epoch 774/1000\n",
      "40/40 - 0s - loss: 0.0063 - 334ms/epoch - 8ms/step\n",
      "Epoch 775/1000\n",
      "40/40 - 0s - loss: 0.0050 - 343ms/epoch - 9ms/step\n",
      "Epoch 776/1000\n",
      "40/40 - 0s - loss: 0.0043 - 350ms/epoch - 9ms/step\n",
      "Epoch 777/1000\n",
      "40/40 - 0s - loss: 0.0038 - 316ms/epoch - 8ms/step\n",
      "Epoch 778/1000\n",
      "40/40 - 0s - loss: 0.0035 - 317ms/epoch - 8ms/step\n",
      "Epoch 779/1000\n",
      "40/40 - 0s - loss: 0.0035 - 331ms/epoch - 8ms/step\n",
      "Epoch 780/1000\n",
      "40/40 - 0s - loss: 0.0032 - 317ms/epoch - 8ms/step\n",
      "Epoch 781/1000\n",
      "40/40 - 0s - loss: 0.0032 - 351ms/epoch - 9ms/step\n",
      "Epoch 782/1000\n",
      "40/40 - 0s - loss: 0.0039 - 322ms/epoch - 8ms/step\n",
      "Epoch 783/1000\n",
      "40/40 - 0s - loss: 0.0036 - 339ms/epoch - 8ms/step\n",
      "Epoch 784/1000\n",
      "40/40 - 0s - loss: 0.0027 - 341ms/epoch - 9ms/step\n",
      "Epoch 785/1000\n",
      "40/40 - 0s - loss: 0.0027 - 310ms/epoch - 8ms/step\n",
      "Epoch 786/1000\n",
      "40/40 - 0s - loss: 0.0026 - 343ms/epoch - 9ms/step\n",
      "Epoch 787/1000\n",
      "40/40 - 0s - loss: 0.0024 - 337ms/epoch - 8ms/step\n",
      "Epoch 788/1000\n",
      "40/40 - 0s - loss: 0.0023 - 338ms/epoch - 8ms/step\n",
      "Epoch 789/1000\n",
      "40/40 - 0s - loss: 0.0023 - 338ms/epoch - 8ms/step\n",
      "Epoch 790/1000\n",
      "40/40 - 0s - loss: 0.0023 - 348ms/epoch - 9ms/step\n",
      "Epoch 791/1000\n",
      "40/40 - 0s - loss: 0.0022 - 343ms/epoch - 9ms/step\n",
      "Epoch 792/1000\n",
      "40/40 - 0s - loss: 0.0022 - 362ms/epoch - 9ms/step\n",
      "Epoch 793/1000\n",
      "40/40 - 0s - loss: 0.0021 - 355ms/epoch - 9ms/step\n",
      "Epoch 794/1000\n",
      "40/40 - 0s - loss: 0.0020 - 365ms/epoch - 9ms/step\n",
      "Epoch 795/1000\n",
      "40/40 - 0s - loss: 0.0019 - 373ms/epoch - 9ms/step\n",
      "Epoch 796/1000\n",
      "40/40 - 0s - loss: 0.0018 - 326ms/epoch - 8ms/step\n",
      "Epoch 797/1000\n",
      "40/40 - 0s - loss: 0.0018 - 313ms/epoch - 8ms/step\n",
      "Epoch 798/1000\n",
      "40/40 - 0s - loss: 0.0022 - 324ms/epoch - 8ms/step\n",
      "Epoch 799/1000\n",
      "40/40 - 0s - loss: 0.0019 - 348ms/epoch - 9ms/step\n",
      "Epoch 800/1000\n",
      "40/40 - 0s - loss: 0.0018 - 322ms/epoch - 8ms/step\n",
      "Epoch 801/1000\n",
      "40/40 - 0s - loss: 0.0017 - 336ms/epoch - 8ms/step\n",
      "Epoch 802/1000\n",
      "40/40 - 0s - loss: 0.0017 - 347ms/epoch - 9ms/step\n",
      "Epoch 803/1000\n",
      "40/40 - 0s - loss: 0.0016 - 327ms/epoch - 8ms/step\n",
      "Epoch 804/1000\n",
      "40/40 - 0s - loss: 0.0017 - 338ms/epoch - 8ms/step\n",
      "Epoch 805/1000\n",
      "40/40 - 0s - loss: 0.0016 - 354ms/epoch - 9ms/step\n",
      "Epoch 806/1000\n",
      "40/40 - 0s - loss: 0.0017 - 342ms/epoch - 9ms/step\n",
      "Epoch 807/1000\n",
      "40/40 - 0s - loss: 0.0019 - 307ms/epoch - 8ms/step\n",
      "Epoch 808/1000\n",
      "40/40 - 0s - loss: 0.0019 - 330ms/epoch - 8ms/step\n",
      "Epoch 809/1000\n",
      "40/40 - 0s - loss: 0.0019 - 335ms/epoch - 8ms/step\n",
      "Epoch 810/1000\n",
      "40/40 - 0s - loss: 0.0016 - 332ms/epoch - 8ms/step\n",
      "Epoch 811/1000\n",
      "40/40 - 0s - loss: 0.0016 - 318ms/epoch - 8ms/step\n",
      "Epoch 812/1000\n",
      "40/40 - 0s - loss: 0.0015 - 339ms/epoch - 8ms/step\n",
      "Epoch 813/1000\n",
      "40/40 - 0s - loss: 0.0015 - 310ms/epoch - 8ms/step\n",
      "Epoch 814/1000\n",
      "40/40 - 0s - loss: 0.0016 - 321ms/epoch - 8ms/step\n",
      "Epoch 815/1000\n",
      "40/40 - 0s - loss: 0.0022 - 325ms/epoch - 8ms/step\n",
      "Epoch 816/1000\n",
      "40/40 - 0s - loss: 0.0017 - 316ms/epoch - 8ms/step\n",
      "Epoch 817/1000\n",
      "40/40 - 0s - loss: 0.0013 - 308ms/epoch - 8ms/step\n",
      "Epoch 818/1000\n",
      "40/40 - 0s - loss: 0.0013 - 322ms/epoch - 8ms/step\n",
      "Epoch 819/1000\n",
      "40/40 - 0s - loss: 0.0014 - 315ms/epoch - 8ms/step\n",
      "Epoch 820/1000\n",
      "40/40 - 0s - loss: 0.0013 - 321ms/epoch - 8ms/step\n",
      "Epoch 821/1000\n",
      "40/40 - 0s - loss: 0.0012 - 336ms/epoch - 8ms/step\n",
      "Epoch 822/1000\n",
      "40/40 - 0s - loss: 0.0015 - 340ms/epoch - 9ms/step\n",
      "Epoch 823/1000\n",
      "40/40 - 0s - loss: 0.0015 - 338ms/epoch - 8ms/step\n",
      "Epoch 824/1000\n",
      "40/40 - 0s - loss: 0.0015 - 335ms/epoch - 8ms/step\n",
      "Epoch 825/1000\n",
      "40/40 - 0s - loss: 0.0017 - 365ms/epoch - 9ms/step\n",
      "Epoch 826/1000\n",
      "40/40 - 0s - loss: 0.0016 - 331ms/epoch - 8ms/step\n",
      "Epoch 827/1000\n",
      "40/40 - 0s - loss: 0.0051 - 359ms/epoch - 9ms/step\n",
      "Epoch 828/1000\n",
      "40/40 - 0s - loss: 0.0073 - 349ms/epoch - 9ms/step\n",
      "Epoch 829/1000\n",
      "40/40 - 0s - loss: 0.0055 - 353ms/epoch - 9ms/step\n",
      "Epoch 830/1000\n",
      "40/40 - 0s - loss: 0.0017 - 337ms/epoch - 8ms/step\n",
      "Epoch 831/1000\n",
      "40/40 - 0s - loss: 0.0014 - 376ms/epoch - 9ms/step\n",
      "Epoch 832/1000\n",
      "40/40 - 0s - loss: 0.0014 - 343ms/epoch - 9ms/step\n",
      "Epoch 833/1000\n",
      "40/40 - 0s - loss: 0.0015 - 344ms/epoch - 9ms/step\n",
      "Epoch 834/1000\n",
      "40/40 - 0s - loss: 0.0019 - 350ms/epoch - 9ms/step\n",
      "Epoch 835/1000\n",
      "40/40 - 0s - loss: 0.0018 - 342ms/epoch - 9ms/step\n",
      "Epoch 836/1000\n",
      "40/40 - 0s - loss: 0.0013 - 369ms/epoch - 9ms/step\n",
      "Epoch 837/1000\n",
      "40/40 - 0s - loss: 0.0013 - 327ms/epoch - 8ms/step\n",
      "Epoch 838/1000\n",
      "40/40 - 0s - loss: 0.0012 - 368ms/epoch - 9ms/step\n",
      "Epoch 839/1000\n",
      "40/40 - 0s - loss: 0.0012 - 442ms/epoch - 11ms/step\n",
      "Epoch 840/1000\n",
      "40/40 - 0s - loss: 0.0011 - 435ms/epoch - 11ms/step\n",
      "Epoch 841/1000\n",
      "40/40 - 0s - loss: 0.0013 - 399ms/epoch - 10ms/step\n",
      "Epoch 842/1000\n",
      "40/40 - 0s - loss: 0.0014 - 378ms/epoch - 9ms/step\n",
      "Epoch 843/1000\n",
      "40/40 - 0s - loss: 0.0011 - 410ms/epoch - 10ms/step\n",
      "Epoch 844/1000\n",
      "40/40 - 0s - loss: 0.0012 - 363ms/epoch - 9ms/step\n",
      "Epoch 845/1000\n",
      "40/40 - 0s - loss: 0.0011 - 426ms/epoch - 11ms/step\n",
      "Epoch 846/1000\n",
      "40/40 - 0s - loss: 0.0071 - 380ms/epoch - 9ms/step\n",
      "Epoch 847/1000\n",
      "40/40 - 0s - loss: 0.0192 - 342ms/epoch - 9ms/step\n",
      "Epoch 848/1000\n",
      "40/40 - 0s - loss: 0.0160 - 392ms/epoch - 10ms/step\n",
      "Epoch 849/1000\n",
      "40/40 - 0s - loss: 0.0042 - 397ms/epoch - 10ms/step\n",
      "Epoch 850/1000\n",
      "40/40 - 0s - loss: 0.0020 - 379ms/epoch - 9ms/step\n",
      "Epoch 851/1000\n",
      "40/40 - 0s - loss: 0.0020 - 333ms/epoch - 8ms/step\n",
      "Epoch 852/1000\n",
      "40/40 - 0s - loss: 0.0017 - 359ms/epoch - 9ms/step\n",
      "Epoch 853/1000\n",
      "40/40 - 0s - loss: 0.0014 - 353ms/epoch - 9ms/step\n",
      "Epoch 854/1000\n",
      "40/40 - 0s - loss: 0.0012 - 360ms/epoch - 9ms/step\n",
      "Epoch 855/1000\n",
      "40/40 - 0s - loss: 0.0012 - 360ms/epoch - 9ms/step\n",
      "Epoch 856/1000\n",
      "40/40 - 0s - loss: 0.0012 - 365ms/epoch - 9ms/step\n",
      "Epoch 857/1000\n",
      "40/40 - 0s - loss: 0.0012 - 344ms/epoch - 9ms/step\n",
      "Epoch 858/1000\n",
      "40/40 - 0s - loss: 0.0011 - 352ms/epoch - 9ms/step\n",
      "Epoch 859/1000\n",
      "40/40 - 0s - loss: 0.0011 - 314ms/epoch - 8ms/step\n",
      "Epoch 860/1000\n",
      "40/40 - 0s - loss: 0.0011 - 332ms/epoch - 8ms/step\n",
      "Epoch 861/1000\n",
      "40/40 - 0s - loss: 0.0010 - 335ms/epoch - 8ms/step\n",
      "Epoch 862/1000\n",
      "40/40 - 0s - loss: 0.0011 - 318ms/epoch - 8ms/step\n",
      "Epoch 863/1000\n",
      "40/40 - 0s - loss: 0.0012 - 347ms/epoch - 9ms/step\n",
      "Epoch 864/1000\n",
      "40/40 - 0s - loss: 0.0012 - 323ms/epoch - 8ms/step\n",
      "Epoch 865/1000\n",
      "40/40 - 0s - loss: 0.0011 - 327ms/epoch - 8ms/step\n",
      "Epoch 866/1000\n",
      "40/40 - 0s - loss: 9.9957e-04 - 341ms/epoch - 9ms/step\n",
      "Epoch 867/1000\n",
      "40/40 - 0s - loss: 0.0012 - 337ms/epoch - 8ms/step\n",
      "Epoch 868/1000\n",
      "40/40 - 0s - loss: 0.0014 - 328ms/epoch - 8ms/step\n",
      "Epoch 869/1000\n",
      "40/40 - 0s - loss: 0.0013 - 339ms/epoch - 8ms/step\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 870/1000\n",
      "40/40 - 0s - loss: 0.0011 - 369ms/epoch - 9ms/step\n",
      "Epoch 871/1000\n",
      "40/40 - 0s - loss: 0.0010 - 330ms/epoch - 8ms/step\n",
      "Epoch 872/1000\n",
      "40/40 - 0s - loss: 0.0013 - 352ms/epoch - 9ms/step\n",
      "Epoch 873/1000\n",
      "40/40 - 0s - loss: 0.0018 - 338ms/epoch - 8ms/step\n",
      "Epoch 874/1000\n",
      "40/40 - 0s - loss: 0.0013 - 330ms/epoch - 8ms/step\n",
      "Epoch 875/1000\n",
      "40/40 - 0s - loss: 0.0013 - 311ms/epoch - 8ms/step\n",
      "Epoch 876/1000\n",
      "40/40 - 0s - loss: 9.9638e-04 - 326ms/epoch - 8ms/step\n",
      "Epoch 877/1000\n",
      "40/40 - 0s - loss: 9.0814e-04 - 332ms/epoch - 8ms/step\n",
      "Epoch 878/1000\n",
      "40/40 - 0s - loss: 8.6120e-04 - 357ms/epoch - 9ms/step\n",
      "Epoch 879/1000\n",
      "40/40 - 0s - loss: 8.6904e-04 - 340ms/epoch - 8ms/step\n",
      "Epoch 880/1000\n",
      "40/40 - 0s - loss: 8.3015e-04 - 335ms/epoch - 8ms/step\n",
      "Epoch 881/1000\n",
      "40/40 - 0s - loss: 0.0053 - 355ms/epoch - 9ms/step\n",
      "Epoch 882/1000\n",
      "40/40 - 0s - loss: 0.0060 - 334ms/epoch - 8ms/step\n",
      "Epoch 883/1000\n",
      "40/40 - 0s - loss: 0.0021 - 369ms/epoch - 9ms/step\n",
      "Epoch 884/1000\n",
      "40/40 - 0s - loss: 0.0012 - 399ms/epoch - 10ms/step\n",
      "Epoch 885/1000\n",
      "40/40 - 0s - loss: 0.0012 - 407ms/epoch - 10ms/step\n",
      "Epoch 886/1000\n",
      "40/40 - 0s - loss: 0.0011 - 370ms/epoch - 9ms/step\n",
      "Epoch 887/1000\n",
      "40/40 - 0s - loss: 0.0013 - 405ms/epoch - 10ms/step\n",
      "Epoch 888/1000\n",
      "40/40 - 0s - loss: 0.0010 - 456ms/epoch - 11ms/step\n",
      "Epoch 889/1000\n",
      "40/40 - 0s - loss: 9.1377e-04 - 377ms/epoch - 9ms/step\n",
      "Epoch 890/1000\n",
      "40/40 - 0s - loss: 8.3260e-04 - 342ms/epoch - 9ms/step\n",
      "Epoch 891/1000\n",
      "40/40 - 0s - loss: 8.8582e-04 - 350ms/epoch - 9ms/step\n",
      "Epoch 892/1000\n",
      "40/40 - 0s - loss: 7.9147e-04 - 356ms/epoch - 9ms/step\n",
      "Epoch 893/1000\n",
      "40/40 - 0s - loss: 8.2580e-04 - 300ms/epoch - 8ms/step\n",
      "Epoch 894/1000\n",
      "40/40 - 0s - loss: 8.0941e-04 - 362ms/epoch - 9ms/step\n",
      "Epoch 895/1000\n",
      "40/40 - 0s - loss: 8.0500e-04 - 333ms/epoch - 8ms/step\n",
      "Epoch 896/1000\n",
      "40/40 - 0s - loss: 0.0013 - 353ms/epoch - 9ms/step\n",
      "Epoch 897/1000\n",
      "40/40 - 0s - loss: 0.0010 - 410ms/epoch - 10ms/step\n",
      "Epoch 898/1000\n",
      "40/40 - 0s - loss: 0.0035 - 391ms/epoch - 10ms/step\n",
      "Epoch 899/1000\n",
      "40/40 - 0s - loss: 0.0032 - 323ms/epoch - 8ms/step\n",
      "Epoch 900/1000\n",
      "40/40 - 0s - loss: 9.9506e-04 - 340ms/epoch - 9ms/step\n",
      "Epoch 901/1000\n",
      "40/40 - 0s - loss: 0.0012 - 355ms/epoch - 9ms/step\n",
      "Epoch 902/1000\n",
      "40/40 - 0s - loss: 9.6159e-04 - 327ms/epoch - 8ms/step\n",
      "Epoch 903/1000\n",
      "40/40 - 0s - loss: 0.0011 - 312ms/epoch - 8ms/step\n",
      "Epoch 904/1000\n",
      "40/40 - 0s - loss: 7.6997e-04 - 375ms/epoch - 9ms/step\n",
      "Epoch 905/1000\n",
      "40/40 - 0s - loss: 7.3856e-04 - 361ms/epoch - 9ms/step\n",
      "Epoch 906/1000\n",
      "40/40 - 0s - loss: 7.4163e-04 - 342ms/epoch - 9ms/step\n",
      "Epoch 907/1000\n",
      "40/40 - 0s - loss: 0.0024 - 324ms/epoch - 8ms/step\n",
      "Epoch 908/1000\n",
      "40/40 - 0s - loss: 0.0016 - 347ms/epoch - 9ms/step\n",
      "Epoch 909/1000\n",
      "40/40 - 0s - loss: 0.0014 - 323ms/epoch - 8ms/step\n",
      "Epoch 910/1000\n",
      "40/40 - 0s - loss: 0.0015 - 315ms/epoch - 8ms/step\n",
      "Epoch 911/1000\n",
      "40/40 - 0s - loss: 9.6629e-04 - 338ms/epoch - 8ms/step\n",
      "Epoch 912/1000\n",
      "40/40 - 0s - loss: 7.7509e-04 - 316ms/epoch - 8ms/step\n",
      "Epoch 913/1000\n",
      "40/40 - 0s - loss: 7.2816e-04 - 342ms/epoch - 9ms/step\n",
      "Epoch 914/1000\n",
      "40/40 - 0s - loss: 6.5932e-04 - 345ms/epoch - 9ms/step\n",
      "Epoch 915/1000\n",
      "40/40 - 0s - loss: 6.9331e-04 - 345ms/epoch - 9ms/step\n",
      "Epoch 916/1000\n",
      "40/40 - 0s - loss: 7.6298e-04 - 325ms/epoch - 8ms/step\n",
      "Epoch 917/1000\n",
      "40/40 - 0s - loss: 0.0016 - 306ms/epoch - 8ms/step\n",
      "Epoch 918/1000\n",
      "40/40 - 0s - loss: 0.0013 - 307ms/epoch - 8ms/step\n",
      "Epoch 919/1000\n",
      "40/40 - 0s - loss: 0.0022 - 324ms/epoch - 8ms/step\n",
      "Epoch 920/1000\n",
      "40/40 - 0s - loss: 0.0013 - 327ms/epoch - 8ms/step\n",
      "Epoch 921/1000\n",
      "40/40 - 0s - loss: 9.0062e-04 - 312ms/epoch - 8ms/step\n",
      "Epoch 922/1000\n",
      "40/40 - 0s - loss: 9.4930e-04 - 323ms/epoch - 8ms/step\n",
      "Epoch 923/1000\n",
      "40/40 - 0s - loss: 6.7815e-04 - 321ms/epoch - 8ms/step\n",
      "Epoch 924/1000\n",
      "40/40 - 0s - loss: 5.2576e-04 - 339ms/epoch - 8ms/step\n",
      "Epoch 925/1000\n",
      "40/40 - 0s - loss: 7.6498e-04 - 329ms/epoch - 8ms/step\n",
      "Epoch 926/1000\n",
      "40/40 - 0s - loss: 6.6478e-04 - 330ms/epoch - 8ms/step\n",
      "Epoch 927/1000\n",
      "40/40 - 0s - loss: 5.5287e-04 - 314ms/epoch - 8ms/step\n",
      "Epoch 928/1000\n",
      "40/40 - 0s - loss: 5.8538e-04 - 364ms/epoch - 9ms/step\n",
      "Epoch 929/1000\n",
      "40/40 - 0s - loss: 6.1307e-04 - 364ms/epoch - 9ms/step\n",
      "Epoch 930/1000\n",
      "40/40 - 0s - loss: 5.6109e-04 - 390ms/epoch - 10ms/step\n",
      "Epoch 931/1000\n",
      "40/40 - 0s - loss: 6.5123e-04 - 359ms/epoch - 9ms/step\n",
      "Epoch 932/1000\n",
      "40/40 - 0s - loss: 5.3985e-04 - 362ms/epoch - 9ms/step\n",
      "Epoch 933/1000\n",
      "40/40 - 0s - loss: 6.6599e-04 - 392ms/epoch - 10ms/step\n",
      "Epoch 934/1000\n",
      "40/40 - 0s - loss: 5.4048e-04 - 351ms/epoch - 9ms/step\n",
      "Epoch 935/1000\n",
      "40/40 - 0s - loss: 5.4709e-04 - 327ms/epoch - 8ms/step\n",
      "Epoch 936/1000\n",
      "40/40 - 0s - loss: 9.8846e-04 - 320ms/epoch - 8ms/step\n",
      "Epoch 937/1000\n",
      "40/40 - 0s - loss: 0.0019 - 340ms/epoch - 8ms/step\n",
      "Epoch 938/1000\n",
      "40/40 - 0s - loss: 0.0024 - 314ms/epoch - 8ms/step\n",
      "Epoch 939/1000\n",
      "40/40 - 0s - loss: 0.0012 - 332ms/epoch - 8ms/step\n",
      "Epoch 940/1000\n",
      "40/40 - 0s - loss: 0.0012 - 332ms/epoch - 8ms/step\n",
      "Epoch 941/1000\n",
      "40/40 - 0s - loss: 8.4323e-04 - 346ms/epoch - 9ms/step\n",
      "Epoch 942/1000\n",
      "40/40 - 0s - loss: 7.3195e-04 - 360ms/epoch - 9ms/step\n",
      "Epoch 943/1000\n",
      "40/40 - 0s - loss: 0.0040 - 345ms/epoch - 9ms/step\n",
      "Epoch 944/1000\n",
      "40/40 - 0s - loss: 0.0223 - 354ms/epoch - 9ms/step\n",
      "Epoch 945/1000\n",
      "40/40 - 0s - loss: 0.0225 - 339ms/epoch - 8ms/step\n",
      "Epoch 946/1000\n",
      "40/40 - 0s - loss: 0.0241 - 330ms/epoch - 8ms/step\n",
      "Epoch 947/1000\n",
      "40/40 - 0s - loss: 0.0182 - 328ms/epoch - 8ms/step\n",
      "Epoch 948/1000\n",
      "40/40 - 0s - loss: 0.0182 - 323ms/epoch - 8ms/step\n",
      "Epoch 949/1000\n",
      "40/40 - 0s - loss: 0.0186 - 338ms/epoch - 8ms/step\n",
      "Epoch 950/1000\n",
      "40/40 - 0s - loss: 0.0212 - 325ms/epoch - 8ms/step\n",
      "Epoch 951/1000\n",
      "40/40 - 0s - loss: 0.0153 - 332ms/epoch - 8ms/step\n",
      "Epoch 952/1000\n",
      "40/40 - 0s - loss: 0.0300 - 350ms/epoch - 9ms/step\n",
      "Epoch 953/1000\n",
      "40/40 - 0s - loss: 0.0191 - 334ms/epoch - 8ms/step\n",
      "Epoch 954/1000\n",
      "40/40 - 0s - loss: 0.0181 - 345ms/epoch - 9ms/step\n",
      "Epoch 955/1000\n",
      "40/40 - 0s - loss: 0.0170 - 341ms/epoch - 9ms/step\n",
      "Epoch 956/1000\n",
      "40/40 - 0s - loss: 0.0157 - 371ms/epoch - 9ms/step\n",
      "Epoch 957/1000\n",
      "40/40 - 0s - loss: 0.0127 - 365ms/epoch - 9ms/step\n",
      "Epoch 958/1000\n",
      "40/40 - 0s - loss: 0.0087 - 309ms/epoch - 8ms/step\n",
      "Epoch 959/1000\n",
      "40/40 - 0s - loss: 0.0145 - 333ms/epoch - 8ms/step\n",
      "Epoch 960/1000\n",
      "40/40 - 0s - loss: 0.0228 - 318ms/epoch - 8ms/step\n",
      "Epoch 961/1000\n",
      "40/40 - 0s - loss: 0.0198 - 338ms/epoch - 8ms/step\n",
      "Epoch 962/1000\n",
      "40/40 - 0s - loss: 0.0171 - 323ms/epoch - 8ms/step\n",
      "Epoch 963/1000\n",
      "40/40 - 0s - loss: 0.0139 - 308ms/epoch - 8ms/step\n",
      "Epoch 964/1000\n",
      "40/40 - 0s - loss: 0.0067 - 312ms/epoch - 8ms/step\n",
      "Epoch 965/1000\n",
      "40/40 - 0s - loss: 0.0127 - 330ms/epoch - 8ms/step\n",
      "Epoch 966/1000\n",
      "40/40 - 0s - loss: 0.0237 - 319ms/epoch - 8ms/step\n",
      "Epoch 967/1000\n",
      "40/40 - 0s - loss: 0.0096 - 323ms/epoch - 8ms/step\n",
      "Epoch 968/1000\n",
      "40/40 - 0s - loss: 0.0052 - 304ms/epoch - 8ms/step\n",
      "Epoch 969/1000\n",
      "40/40 - 0s - loss: 0.0081 - 308ms/epoch - 8ms/step\n",
      "Epoch 970/1000\n",
      "40/40 - 0s - loss: 0.0065 - 334ms/epoch - 8ms/step\n",
      "Epoch 971/1000\n",
      "40/40 - 0s - loss: 0.0032 - 326ms/epoch - 8ms/step\n",
      "Epoch 972/1000\n",
      "40/40 - 0s - loss: 0.0026 - 321ms/epoch - 8ms/step\n",
      "Epoch 973/1000\n",
      "40/40 - 0s - loss: 0.0023 - 339ms/epoch - 8ms/step\n",
      "Epoch 974/1000\n",
      "40/40 - 0s - loss: 0.0032 - 321ms/epoch - 8ms/step\n",
      "Epoch 975/1000\n",
      "40/40 - 0s - loss: 0.0055 - 310ms/epoch - 8ms/step\n",
      "Epoch 976/1000\n",
      "40/40 - 0s - loss: 0.0023 - 358ms/epoch - 9ms/step\n",
      "Epoch 977/1000\n",
      "40/40 - 0s - loss: 0.0018 - 355ms/epoch - 9ms/step\n",
      "Epoch 978/1000\n",
      "40/40 - 0s - loss: 0.0015 - 344ms/epoch - 9ms/step\n",
      "Epoch 979/1000\n",
      "40/40 - 0s - loss: 0.0013 - 374ms/epoch - 9ms/step\n",
      "Epoch 980/1000\n",
      "40/40 - 0s - loss: 0.0014 - 368ms/epoch - 9ms/step\n",
      "Epoch 981/1000\n",
      "40/40 - 0s - loss: 0.0013 - 365ms/epoch - 9ms/step\n",
      "Epoch 982/1000\n",
      "40/40 - 0s - loss: 0.0013 - 346ms/epoch - 9ms/step\n",
      "Epoch 983/1000\n",
      "40/40 - 0s - loss: 0.0013 - 335ms/epoch - 8ms/step\n",
      "Epoch 984/1000\n",
      "40/40 - 0s - loss: 0.0011 - 333ms/epoch - 8ms/step\n",
      "Epoch 985/1000\n",
      "40/40 - 0s - loss: 0.0017 - 327ms/epoch - 8ms/step\n",
      "Epoch 986/1000\n",
      "40/40 - 0s - loss: 0.0012 - 319ms/epoch - 8ms/step\n",
      "Epoch 987/1000\n",
      "40/40 - 0s - loss: 9.6086e-04 - 318ms/epoch - 8ms/step\n",
      "Epoch 988/1000\n",
      "40/40 - 0s - loss: 0.0010 - 342ms/epoch - 9ms/step\n",
      "Epoch 989/1000\n",
      "40/40 - 0s - loss: 8.8949e-04 - 344ms/epoch - 9ms/step\n",
      "Epoch 990/1000\n",
      "40/40 - 0s - loss: 9.5853e-04 - 328ms/epoch - 8ms/step\n",
      "Epoch 991/1000\n",
      "40/40 - 0s - loss: 9.8523e-04 - 340ms/epoch - 9ms/step\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 992/1000\n",
      "40/40 - 0s - loss: 0.0011 - 314ms/epoch - 8ms/step\n",
      "Epoch 993/1000\n",
      "40/40 - 0s - loss: 0.0014 - 318ms/epoch - 8ms/step\n",
      "Epoch 994/1000\n",
      "40/40 - 0s - loss: 8.8786e-04 - 335ms/epoch - 8ms/step\n",
      "Epoch 995/1000\n",
      "40/40 - 0s - loss: 0.0010 - 314ms/epoch - 8ms/step\n",
      "Epoch 996/1000\n",
      "40/40 - 0s - loss: 8.7613e-04 - 322ms/epoch - 8ms/step\n",
      "Epoch 997/1000\n",
      "40/40 - 0s - loss: 0.0013 - 349ms/epoch - 9ms/step\n",
      "Epoch 998/1000\n",
      "40/40 - 0s - loss: 0.0011 - 334ms/epoch - 8ms/step\n",
      "Epoch 999/1000\n",
      "40/40 - 0s - loss: 9.1230e-04 - 316ms/epoch - 8ms/step\n",
      "Epoch 1000/1000\n",
      "40/40 - 0s - loss: 9.1556e-04 - 314ms/epoch - 8ms/step\n"
     ]
    }
   ],
   "source": [
    "with tf.device('/gpu:0'):\n",
    "    start = dt.now()\n",
    "    model.fit(XTrain_s,np.array(TTrain),epochs=noEpochs,verbose=2)\n",
    "    trainTime=(dt.now() - start)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "a852559c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train time:  0:05:50.485065 s\n"
     ]
    }
   ],
   "source": [
    "print(\"Train time: \",trainTime,\"s\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e24d20bf",
   "metadata": {},
   "source": [
    "## Inference"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "5df4584d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10/10 [==============================] - 0s 10ms/step\n"
     ]
    }
   ],
   "source": [
    "with tf.device('/cpu:0'):\n",
    "    Y_hat=model.predict(XTest_s)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "0edfc195",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1278, 50, 4)\n",
      "(1278,)\n"
     ]
    }
   ],
   "source": [
    "print(XTrain_s.shape)\n",
    "print(TTrain.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d28eda96",
   "metadata": {},
   "source": [
    "## Plot prediction "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "e0d91afc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2520x1080 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "from matplotlib.pyplot import figure\n",
    "plt.figure(figsize=(35,15))\n",
    "plt.plot(TTest,color='blue',label='Actual',alpha=0.5,linewidth=7.0)\n",
    "plt.plot(Y_hat,color='red',label='Predicted')\n",
    "plt.xlabel('Number')\n",
    "plt.ylabel('VO2')\n",
    "plt.title('Actual vs Predicted')\n",
    "plt.tight_layout()\n",
    "plt.legend(prop={'size': 30})\n",
    "plt.savefig(figPredictName,bbox_inches='tight',dpi=150)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6f8ef8bc",
   "metadata": {},
   "source": [
    "## Save model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "95627e67",
   "metadata": {},
   "outputs": [],
   "source": [
    "if savedModel:\n",
    "    model.save(modelSaveName)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "7441ce70",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1476x756 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Y_hat=np.transpose(Y_hat)\n",
    "TTest=TTest.reshape(len(TTest),1)\n",
    "#print(Y_hat)\n",
    "#diff=np.concatenate((Y_hat,TTest),axis=1)\n",
    "diff=np.subtract(Y_hat,TTest)\n",
    "#print(diff)\n",
    "\n",
    "plt.figure(figsize=(20.5,15.5))\n",
    "plt.plot(diff,color='red')\n",
    "plt.xlabel('Number')\n",
    "plt.ylabel('Diff')\n",
    "plt.title('Difference between grounded and predicted value')\n",
    "fig = plt.gcf()\n",
    "fig.set_size_inches(20.5, 10.5)\n",
    "#fig.savefig('1st-stage-subj-1_L'+results+'.jpg', dpi=150)\n",
    "plt.show()\n",
    "#model.save('1st-stage-subj-1.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "83113186",
   "metadata": {},
   "outputs": [],
   "source": [
    "# print(TTest.shape)\n",
    "# print(XTest.shape)\n",
    "# print(XTrain.shape)\n",
    "# print(TTrain.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "2f475c68",
   "metadata": {},
   "outputs": [],
   "source": [
    "# print(TTest)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "5d4a8537",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[ 1.85989285  1.5625      0.29739285  1.71119642]\n",
      " [ 2.62532902  4.075      -1.44967098  3.35016451]\n",
      " [ 3.65182638  4.075      -0.42317362  3.86341319]\n",
      " ...\n",
      " [40.20728302 39.97        0.23728302 40.08864151]\n",
      " [39.41952896 39.97       -0.55047104 39.69476448]\n",
      " [40.28229523 40.222       0.06029523 40.25214761]]\n"
     ]
    }
   ],
   "source": [
    "res=np.concatenate((Y_hat,TTest,Y_hat-TTest,(Y_hat+TTest)/2),axis=1)\n",
    "print(res)\n",
    "\n",
    "if exportCSV:\n",
    "    pd.DataFrame(res).to_csv('Subj-'+subjectID+'-predict.csv')\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}