009-Transfer-Learning-DosaNoDosa.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Dosa / No Dosa - Transfer Learning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Frame the Problem"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We want to classify the images in two categories - **Dosa** and **No-Dosa**. This is a classical two-class image classification problem.\n",
    "- Dosa\n",
    "- No-Dosa"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Acquire the Data\n",
    "\n",
    "These images very downloaded google.com/images with \"reuse with modification\" filter to allow for commercial reuse."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "#!mkdir data\n",
    "#!wget https://bit.do/dosa-nodosa -O data/food-binary.zip\n",
    "#!cd data && unzip food-binary.zip"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load basic libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import keras\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "import vis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Transfer Learning with Image Augementation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Step 1: Prepare the images and labels**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.preprocessing.image import ImageDataGenerator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = ImageDataGenerator(validation_split=0.2, \n",
    "                          rescale=1/255,\n",
    "                          samplewise_center=False, # set input mean to 0 over the sample\n",
    "                          samplewise_std_normalization=False,  # divide inputs by std of the sample\n",
    "                          rotation_range=90,       # randomly rotate images in the range (degrees, 0 to 180)\n",
    "                          width_shift_range=0.2,   # randomly shift images horizontally (fraction of total width)\n",
    "                          height_shift_range=0.2,  # randomly shift images vertically (fraction of total height)\n",
    "                          fill_mode='reflect',     # filling the area outside\n",
    "                          zoom_range=0.4,          # random zoom\n",
    "                          horizontal_flip=True,    # randomly flip images\n",
    "                          vertical_flip=True)      # randomly flip images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 260 images belonging to 2 classes.\n"
     ]
    }
   ],
   "source": [
    "train_generator = data.flow_from_directory(directory=\"data/food-binary/\", \n",
    "                                           target_size=(224,224),\n",
    "                                           class_mode=\"categorical\", \n",
    "                                           shuffle=True, \n",
    "                                           batch_size=8, \n",
    "                                           subset='training')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 65 images belonging to 2 classes.\n"
     ]
    }
   ],
   "source": [
    "val_generator = data.flow_from_directory(directory=\"data/food-binary/\", \n",
    "                                           target_size=(224,224),\n",
    "                                           class_mode=\"categorical\", \n",
    "                                           shuffle=True, \n",
    "                                           batch_size=8, \n",
    "                                           subset='validation')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Step 2: Using a Transfer Learning Model **"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let us load the ResNet50 model without the classifier layer and **do not** train the weights"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.applications import ResNet50\n",
    "from tensorflow.keras.models import Model\n",
    "from tensorflow.keras.layers import Conv2D, MaxPooling2D, Dropout, Dense, Flatten, GlobalAveragePooling2D"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "conv_base = ResNet50(include_top=False, weights='imagenet', input_shape=(224,224,3), pooling=\"avg\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "for layer in conv_base.layers:\n",
    "    layer.trainable = False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "x = conv_base.output\n",
    "x = Dense(512, activation='relu')(x) # let's add a fully-connected layer\n",
    "x = Dropout(0.25)(x)\n",
    "x = Dense(128, activation='relu')(x) # let's add a fully-connected layer\n",
    "x = Dropout(0.25)(x)\n",
    "predictions = Dense(2, activation='softmax')(x)\n",
    "\n",
    "# this is the model we will train\n",
    "m = Model(inputs=conv_base.input, outputs=predictions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "input_1 (InputLayer)            [(None, 224, 224, 3) 0                                            \n",
      "__________________________________________________________________________________________________\n",
      "conv1_pad (ZeroPadding2D)       (None, 230, 230, 3)  0           input_1[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "conv1 (Conv2D)                  (None, 112, 112, 64) 9472        conv1_pad[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "bn_conv1 (BatchNormalizationV1) (None, 112, 112, 64) 256         conv1[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "activation (Activation)         (None, 112, 112, 64) 0           bn_conv1[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "pool1_pad (ZeroPadding2D)       (None, 114, 114, 64) 0           activation[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "max_pooling2d (MaxPooling2D)    (None, 56, 56, 64)   0           pool1_pad[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "res2a_branch2a (Conv2D)         (None, 56, 56, 64)   4160        max_pooling2d[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn2a_branch2a (BatchNormalizati (None, 56, 56, 64)   256         res2a_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_1 (Activation)       (None, 56, 56, 64)   0           bn2a_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res2a_branch2b (Conv2D)         (None, 56, 56, 64)   36928       activation_1[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "bn2a_branch2b (BatchNormalizati (None, 56, 56, 64)   256         res2a_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_2 (Activation)       (None, 56, 56, 64)   0           bn2a_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res2a_branch2c (Conv2D)         (None, 56, 56, 256)  16640       activation_2[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "res2a_branch1 (Conv2D)          (None, 56, 56, 256)  16640       max_pooling2d[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn2a_branch2c (BatchNormalizati (None, 56, 56, 256)  1024        res2a_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "bn2a_branch1 (BatchNormalizatio (None, 56, 56, 256)  1024        res2a_branch1[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "add (Add)                       (None, 56, 56, 256)  0           bn2a_branch2c[0][0]              \n",
      "                                                                 bn2a_branch1[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "activation_3 (Activation)       (None, 56, 56, 256)  0           add[0][0]                        \n",
      "__________________________________________________________________________________________________\n",
      "res2b_branch2a (Conv2D)         (None, 56, 56, 64)   16448       activation_3[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "bn2b_branch2a (BatchNormalizati (None, 56, 56, 64)   256         res2b_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_4 (Activation)       (None, 56, 56, 64)   0           bn2b_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res2b_branch2b (Conv2D)         (None, 56, 56, 64)   36928       activation_4[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "bn2b_branch2b (BatchNormalizati (None, 56, 56, 64)   256         res2b_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_5 (Activation)       (None, 56, 56, 64)   0           bn2b_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res2b_branch2c (Conv2D)         (None, 56, 56, 256)  16640       activation_5[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "bn2b_branch2c (BatchNormalizati (None, 56, 56, 256)  1024        res2b_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "add_1 (Add)                     (None, 56, 56, 256)  0           bn2b_branch2c[0][0]              \n",
      "                                                                 activation_3[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "activation_6 (Activation)       (None, 56, 56, 256)  0           add_1[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "res2c_branch2a (Conv2D)         (None, 56, 56, 64)   16448       activation_6[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "bn2c_branch2a (BatchNormalizati (None, 56, 56, 64)   256         res2c_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_7 (Activation)       (None, 56, 56, 64)   0           bn2c_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res2c_branch2b (Conv2D)         (None, 56, 56, 64)   36928       activation_7[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "bn2c_branch2b (BatchNormalizati (None, 56, 56, 64)   256         res2c_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_8 (Activation)       (None, 56, 56, 64)   0           bn2c_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res2c_branch2c (Conv2D)         (None, 56, 56, 256)  16640       activation_8[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "bn2c_branch2c (BatchNormalizati (None, 56, 56, 256)  1024        res2c_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "add_2 (Add)                     (None, 56, 56, 256)  0           bn2c_branch2c[0][0]              \n",
      "                                                                 activation_6[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "activation_9 (Activation)       (None, 56, 56, 256)  0           add_2[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "res3a_branch2a (Conv2D)         (None, 28, 28, 128)  32896       activation_9[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "bn3a_branch2a (BatchNormalizati (None, 28, 28, 128)  512         res3a_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_10 (Activation)      (None, 28, 28, 128)  0           bn3a_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res3a_branch2b (Conv2D)         (None, 28, 28, 128)  147584      activation_10[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn3a_branch2b (BatchNormalizati (None, 28, 28, 128)  512         res3a_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_11 (Activation)      (None, 28, 28, 128)  0           bn3a_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res3a_branch2c (Conv2D)         (None, 28, 28, 512)  66048       activation_11[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res3a_branch1 (Conv2D)          (None, 28, 28, 512)  131584      activation_9[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "bn3a_branch2c (BatchNormalizati (None, 28, 28, 512)  2048        res3a_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "bn3a_branch1 (BatchNormalizatio (None, 28, 28, 512)  2048        res3a_branch1[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "add_3 (Add)                     (None, 28, 28, 512)  0           bn3a_branch2c[0][0]              \n",
      "                                                                 bn3a_branch1[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "activation_12 (Activation)      (None, 28, 28, 512)  0           add_3[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "res3b_branch2a (Conv2D)         (None, 28, 28, 128)  65664       activation_12[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn3b_branch2a (BatchNormalizati (None, 28, 28, 128)  512         res3b_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_13 (Activation)      (None, 28, 28, 128)  0           bn3b_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res3b_branch2b (Conv2D)         (None, 28, 28, 128)  147584      activation_13[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn3b_branch2b (BatchNormalizati (None, 28, 28, 128)  512         res3b_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_14 (Activation)      (None, 28, 28, 128)  0           bn3b_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res3b_branch2c (Conv2D)         (None, 28, 28, 512)  66048       activation_14[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn3b_branch2c (BatchNormalizati (None, 28, 28, 512)  2048        res3b_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "add_4 (Add)                     (None, 28, 28, 512)  0           bn3b_branch2c[0][0]              \n",
      "                                                                 activation_12[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "activation_15 (Activation)      (None, 28, 28, 512)  0           add_4[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "res3c_branch2a (Conv2D)         (None, 28, 28, 128)  65664       activation_15[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn3c_branch2a (BatchNormalizati (None, 28, 28, 128)  512         res3c_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_16 (Activation)      (None, 28, 28, 128)  0           bn3c_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res3c_branch2b (Conv2D)         (None, 28, 28, 128)  147584      activation_16[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn3c_branch2b (BatchNormalizati (None, 28, 28, 128)  512         res3c_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_17 (Activation)      (None, 28, 28, 128)  0           bn3c_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res3c_branch2c (Conv2D)         (None, 28, 28, 512)  66048       activation_17[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn3c_branch2c (BatchNormalizati (None, 28, 28, 512)  2048        res3c_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "add_5 (Add)                     (None, 28, 28, 512)  0           bn3c_branch2c[0][0]              \n",
      "                                                                 activation_15[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "activation_18 (Activation)      (None, 28, 28, 512)  0           add_5[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "res3d_branch2a (Conv2D)         (None, 28, 28, 128)  65664       activation_18[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn3d_branch2a (BatchNormalizati (None, 28, 28, 128)  512         res3d_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_19 (Activation)      (None, 28, 28, 128)  0           bn3d_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res3d_branch2b (Conv2D)         (None, 28, 28, 128)  147584      activation_19[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn3d_branch2b (BatchNormalizati (None, 28, 28, 128)  512         res3d_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_20 (Activation)      (None, 28, 28, 128)  0           bn3d_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res3d_branch2c (Conv2D)         (None, 28, 28, 512)  66048       activation_20[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn3d_branch2c (BatchNormalizati (None, 28, 28, 512)  2048        res3d_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "add_6 (Add)                     (None, 28, 28, 512)  0           bn3d_branch2c[0][0]              \n",
      "                                                                 activation_18[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "activation_21 (Activation)      (None, 28, 28, 512)  0           add_6[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "res4a_branch2a (Conv2D)         (None, 14, 14, 256)  131328      activation_21[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4a_branch2a (BatchNormalizati (None, 14, 14, 256)  1024        res4a_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_22 (Activation)      (None, 14, 14, 256)  0           bn4a_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res4a_branch2b (Conv2D)         (None, 14, 14, 256)  590080      activation_22[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4a_branch2b (BatchNormalizati (None, 14, 14, 256)  1024        res4a_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_23 (Activation)      (None, 14, 14, 256)  0           bn4a_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res4a_branch2c (Conv2D)         (None, 14, 14, 1024) 263168      activation_23[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res4a_branch1 (Conv2D)          (None, 14, 14, 1024) 525312      activation_21[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4a_branch2c (BatchNormalizati (None, 14, 14, 1024) 4096        res4a_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "bn4a_branch1 (BatchNormalizatio (None, 14, 14, 1024) 4096        res4a_branch1[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "add_7 (Add)                     (None, 14, 14, 1024) 0           bn4a_branch2c[0][0]              \n",
      "                                                                 bn4a_branch1[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "activation_24 (Activation)      (None, 14, 14, 1024) 0           add_7[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "res4b_branch2a (Conv2D)         (None, 14, 14, 256)  262400      activation_24[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4b_branch2a (BatchNormalizati (None, 14, 14, 256)  1024        res4b_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_25 (Activation)      (None, 14, 14, 256)  0           bn4b_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res4b_branch2b (Conv2D)         (None, 14, 14, 256)  590080      activation_25[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4b_branch2b (BatchNormalizati (None, 14, 14, 256)  1024        res4b_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_26 (Activation)      (None, 14, 14, 256)  0           bn4b_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res4b_branch2c (Conv2D)         (None, 14, 14, 1024) 263168      activation_26[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4b_branch2c (BatchNormalizati (None, 14, 14, 1024) 4096        res4b_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "add_8 (Add)                     (None, 14, 14, 1024) 0           bn4b_branch2c[0][0]              \n",
      "                                                                 activation_24[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "activation_27 (Activation)      (None, 14, 14, 1024) 0           add_8[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "res4c_branch2a (Conv2D)         (None, 14, 14, 256)  262400      activation_27[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4c_branch2a (BatchNormalizati (None, 14, 14, 256)  1024        res4c_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_28 (Activation)      (None, 14, 14, 256)  0           bn4c_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res4c_branch2b (Conv2D)         (None, 14, 14, 256)  590080      activation_28[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4c_branch2b (BatchNormalizati (None, 14, 14, 256)  1024        res4c_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_29 (Activation)      (None, 14, 14, 256)  0           bn4c_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res4c_branch2c (Conv2D)         (None, 14, 14, 1024) 263168      activation_29[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4c_branch2c (BatchNormalizati (None, 14, 14, 1024) 4096        res4c_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "add_9 (Add)                     (None, 14, 14, 1024) 0           bn4c_branch2c[0][0]              \n",
      "                                                                 activation_27[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "activation_30 (Activation)      (None, 14, 14, 1024) 0           add_9[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "res4d_branch2a (Conv2D)         (None, 14, 14, 256)  262400      activation_30[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4d_branch2a (BatchNormalizati (None, 14, 14, 256)  1024        res4d_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_31 (Activation)      (None, 14, 14, 256)  0           bn4d_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res4d_branch2b (Conv2D)         (None, 14, 14, 256)  590080      activation_31[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4d_branch2b (BatchNormalizati (None, 14, 14, 256)  1024        res4d_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_32 (Activation)      (None, 14, 14, 256)  0           bn4d_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res4d_branch2c (Conv2D)         (None, 14, 14, 1024) 263168      activation_32[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4d_branch2c (BatchNormalizati (None, 14, 14, 1024) 4096        res4d_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "add_10 (Add)                    (None, 14, 14, 1024) 0           bn4d_branch2c[0][0]              \n",
      "                                                                 activation_30[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "activation_33 (Activation)      (None, 14, 14, 1024) 0           add_10[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "res4e_branch2a (Conv2D)         (None, 14, 14, 256)  262400      activation_33[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4e_branch2a (BatchNormalizati (None, 14, 14, 256)  1024        res4e_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_34 (Activation)      (None, 14, 14, 256)  0           bn4e_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res4e_branch2b (Conv2D)         (None, 14, 14, 256)  590080      activation_34[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4e_branch2b (BatchNormalizati (None, 14, 14, 256)  1024        res4e_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_35 (Activation)      (None, 14, 14, 256)  0           bn4e_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res4e_branch2c (Conv2D)         (None, 14, 14, 1024) 263168      activation_35[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4e_branch2c (BatchNormalizati (None, 14, 14, 1024) 4096        res4e_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "add_11 (Add)                    (None, 14, 14, 1024) 0           bn4e_branch2c[0][0]              \n",
      "                                                                 activation_33[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "activation_36 (Activation)      (None, 14, 14, 1024) 0           add_11[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "res4f_branch2a (Conv2D)         (None, 14, 14, 256)  262400      activation_36[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4f_branch2a (BatchNormalizati (None, 14, 14, 256)  1024        res4f_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_37 (Activation)      (None, 14, 14, 256)  0           bn4f_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res4f_branch2b (Conv2D)         (None, 14, 14, 256)  590080      activation_37[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4f_branch2b (BatchNormalizati (None, 14, 14, 256)  1024        res4f_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_38 (Activation)      (None, 14, 14, 256)  0           bn4f_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res4f_branch2c (Conv2D)         (None, 14, 14, 1024) 263168      activation_38[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn4f_branch2c (BatchNormalizati (None, 14, 14, 1024) 4096        res4f_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "add_12 (Add)                    (None, 14, 14, 1024) 0           bn4f_branch2c[0][0]              \n",
      "                                                                 activation_36[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "activation_39 (Activation)      (None, 14, 14, 1024) 0           add_12[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "res5a_branch2a (Conv2D)         (None, 7, 7, 512)    524800      activation_39[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn5a_branch2a (BatchNormalizati (None, 7, 7, 512)    2048        res5a_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_40 (Activation)      (None, 7, 7, 512)    0           bn5a_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res5a_branch2b (Conv2D)         (None, 7, 7, 512)    2359808     activation_40[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn5a_branch2b (BatchNormalizati (None, 7, 7, 512)    2048        res5a_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_41 (Activation)      (None, 7, 7, 512)    0           bn5a_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res5a_branch2c (Conv2D)         (None, 7, 7, 2048)   1050624     activation_41[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res5a_branch1 (Conv2D)          (None, 7, 7, 2048)   2099200     activation_39[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn5a_branch2c (BatchNormalizati (None, 7, 7, 2048)   8192        res5a_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "bn5a_branch1 (BatchNormalizatio (None, 7, 7, 2048)   8192        res5a_branch1[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "add_13 (Add)                    (None, 7, 7, 2048)   0           bn5a_branch2c[0][0]              \n",
      "                                                                 bn5a_branch1[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "activation_42 (Activation)      (None, 7, 7, 2048)   0           add_13[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "res5b_branch2a (Conv2D)         (None, 7, 7, 512)    1049088     activation_42[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn5b_branch2a (BatchNormalizati (None, 7, 7, 512)    2048        res5b_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_43 (Activation)      (None, 7, 7, 512)    0           bn5b_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res5b_branch2b (Conv2D)         (None, 7, 7, 512)    2359808     activation_43[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn5b_branch2b (BatchNormalizati (None, 7, 7, 512)    2048        res5b_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_44 (Activation)      (None, 7, 7, 512)    0           bn5b_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res5b_branch2c (Conv2D)         (None, 7, 7, 2048)   1050624     activation_44[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn5b_branch2c (BatchNormalizati (None, 7, 7, 2048)   8192        res5b_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "add_14 (Add)                    (None, 7, 7, 2048)   0           bn5b_branch2c[0][0]              \n",
      "                                                                 activation_42[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "activation_45 (Activation)      (None, 7, 7, 2048)   0           add_14[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "res5c_branch2a (Conv2D)         (None, 7, 7, 512)    1049088     activation_45[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn5c_branch2a (BatchNormalizati (None, 7, 7, 512)    2048        res5c_branch2a[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_46 (Activation)      (None, 7, 7, 512)    0           bn5c_branch2a[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res5c_branch2b (Conv2D)         (None, 7, 7, 512)    2359808     activation_46[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn5c_branch2b (BatchNormalizati (None, 7, 7, 512)    2048        res5c_branch2b[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_47 (Activation)      (None, 7, 7, 512)    0           bn5c_branch2b[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "res5c_branch2c (Conv2D)         (None, 7, 7, 2048)   1050624     activation_47[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "bn5c_branch2c (BatchNormalizati (None, 7, 7, 2048)   8192        res5c_branch2c[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "add_15 (Add)                    (None, 7, 7, 2048)   0           bn5c_branch2c[0][0]              \n",
      "                                                                 activation_45[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "activation_48 (Activation)      (None, 7, 7, 2048)   0           add_15[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "global_average_pooling2d (Globa (None, 2048)         0           activation_48[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "dense (Dense)                   (None, 512)          1049088     global_average_pooling2d[0][0]   \n",
      "__________________________________________________________________________________________________\n",
      "dropout (Dropout)               (None, 512)          0           dense[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "dense_1 (Dense)                 (None, 128)          65664       dropout[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "dropout_1 (Dropout)             (None, 128)          0           dense_1[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "dense_2 (Dense)                 (None, 2)            258         dropout_1[0][0]                  \n",
      "==================================================================================================\n",
      "Total params: 24,702,722\n",
      "Trainable params: 1,115,010\n",
      "Non-trainable params: 23,587,712\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "m.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "m.compile(loss='binary_crossentropy', optimizer=\"sgd\", metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/8\n",
      "125/125 [==============================] - 33s 261ms/step - loss: 0.5498 - acc: 0.7380 - val_loss: 0.7019 - val_acc: 0.3309\n",
      "Epoch 2/8\n",
      "125/125 [==============================] - 31s 246ms/step - loss: 0.3690 - acc: 0.8480 - val_loss: 0.7806 - val_acc: 0.3217\n",
      "Epoch 3/8\n",
      "125/125 [==============================] - 32s 254ms/step - loss: 0.3469 - acc: 0.8640 - val_loss: 0.7919 - val_acc: 0.3180\n",
      "Epoch 4/8\n",
      "125/125 [==============================] - 30s 241ms/step - loss: 0.3159 - acc: 0.8740 - val_loss: 0.8040 - val_acc: 0.3254\n",
      "Epoch 5/8\n",
      "125/125 [==============================] - 30s 241ms/step - loss: 0.2543 - acc: 0.8920 - val_loss: 1.0224 - val_acc: 0.3272\n",
      "Epoch 6/8\n",
      "117/125 [===========================>..] - ETA: 1s - loss: 0.2511 - acc: 0.8793"
     ]
    }
   ],
   "source": [
    "m_output =  m.fit_generator(train_generator, epochs=8, steps_per_epoch=1000 // 8,\n",
    "                            validation_data=val_generator, validation_steps=600 // 8,\n",
    "                            verbose = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "m_metrics = m_output.history"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.vegalite.v2+json": {
       "$schema": "https://vega.github.io/schema/vega-lite/v2.4.3.json",
       "config": {
        "view": {
         "height": 300,
         "width": 400
        }
       },
       "data": {
        "values": [
         {
          "batch": 1,
          "metrics": "loss",
          "type": "validation",
          "value": 0.701902575133478
         },
         {
          "batch": 2,
          "metrics": "loss",
          "type": "validation",
          "value": 0.7805887259104672
         },
         {
          "batch": 3,
          "metrics": "loss",
          "type": "validation",
          "value": 0.7918856051914832
         },
         {
          "batch": 4,
          "metrics": "loss",
          "type": "validation",
          "value": 0.8039727710077867
         },
         {
          "batch": 5,
          "metrics": "loss",
          "type": "validation",
          "value": 1.0223659182186513
         },
         {
          "batch": 6,
          "metrics": "loss",
          "type": "validation",
          "value": 0.9787187960875385
         },
         {
          "batch": 7,
          "metrics": "loss",
          "type": "validation",
          "value": 1.0965236755197538
         },
         {
          "batch": 8,
          "metrics": "loss",
          "type": "validation",
          "value": 1.0119539955083061
         },
         {
          "batch": 1,
          "metrics": "acc",
          "type": "validation",
          "value": 0.33088235294117646
         },
         {
          "batch": 2,
          "metrics": "acc",
          "type": "validation",
          "value": 0.32169117647058826
         },
         {
          "batch": 3,
          "metrics": "acc",
          "type": "validation",
          "value": 0.3180147058823529
         },
         {
          "batch": 4,
          "metrics": "acc",
          "type": "validation",
          "value": 0.32536764705882354
         },
         {
          "batch": 5,
          "metrics": "acc",
          "type": "validation",
          "value": 0.3272058823529412
         },
         {
          "batch": 6,
          "metrics": "acc",
          "type": "validation",
          "value": 0.3180147058823529
         },
         {
          "batch": 7,
          "metrics": "acc",
          "type": "validation",
          "value": 0.32536764705882354
         },
         {
          "batch": 8,
          "metrics": "acc",
          "type": "validation",
          "value": 0.3272058823529412
         },
         {
          "batch": 1,
          "metrics": "loss",
          "type": "training",
          "value": 0.5502554244869123
         },
         {
          "batch": 2,
          "metrics": "loss",
          "type": "training",
          "value": 0.36763376330699377
         },
         {
          "batch": 3,
          "metrics": "loss",
          "type": "training",
          "value": 0.34893063513132244
         },
         {
          "batch": 4,
          "metrics": "loss",
          "type": "training",
          "value": 0.30799114149881573
         },
         {
          "batch": 5,
          "metrics": "loss",
          "type": "training",
          "value": 0.25493125496356467
         },
         {
          "batch": 6,
          "metrics": "loss",
          "type": "training",
          "value": 0.2680918022429919
         },
         {
          "batch": 7,
          "metrics": "loss",
          "type": "training",
          "value": 0.20834400435924771
         },
         {
          "batch": 8,
          "metrics": "loss",
          "type": "training",
          "value": 0.22445701886918623
         },
         {
          "batch": 1,
          "metrics": "acc",
          "type": "training",
          "value": 0.7378048780487805
         },
         {
          "batch": 2,
          "metrics": "acc",
          "type": "training",
          "value": 0.8485772357723578
         },
         {
          "batch": 3,
          "metrics": "acc",
          "type": "training",
          "value": 0.8633603238866396
         },
         {
          "batch": 4,
          "metrics": "acc",
          "type": "training",
          "value": 0.8780487804878049
         },
         {
          "batch": 5,
          "metrics": "acc",
          "type": "training",
          "value": 0.8922764227642277
         },
         {
          "batch": 6,
          "metrics": "acc",
          "type": "training",
          "value": 0.8780487804878049
         },
         {
          "batch": 7,
          "metrics": "acc",
          "type": "training",
          "value": 0.9136178861788617
         },
         {
          "batch": 8,
          "metrics": "acc",
          "type": "training",
          "value": 0.9078947368421053
         }
        ]
       },
       "facet": {
        "column": {
         "field": "metrics",
         "type": "nominal"
        }
       },
       "resolve": {
        "scale": {
         "color": "shared",
         "x": "shared",
         "y": "independent"
        }
       },
       "spec": {
        "layer": [
         {
          "encoding": {
           "color": {
            "field": "type",
            "type": "nominal"
           },
           "tooltip": [
            {
             "field": "batch",
             "type": "quantitative"
            },
            {
             "field": "value",
             "type": "quantitative"
            }
           ],
           "x": {
            "field": "batch",
            "type": "quantitative"
           },
           "y": {
            "field": "value",
            "type": "quantitative"
           }
          },
          "height": 300,
          "mark": {
           "size": 50,
           "type": "circle"
          },
          "width": 300
         },
         {
          "encoding": {
           "color": {
            "field": "type",
            "type": "nominal"
           },
           "x": {
            "field": "batch",
            "type": "quantitative"
           },
           "y": {
            "field": "value",
            "type": "quantitative"
           }
          },
          "height": 300,
          "mark": "line",
          "width": 300
         }
        ]
       }
      },
      "image/png": "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",
      "text/plain": [
       "<VegaLite 2 object>\n",
       "\n",
       "If you see this message, it means the renderer has not been properly enabled\n",
       "for the frontend that you are using. For more information, see\n",
       "https://altair-viz.github.io/user_guide/troubleshooting.html\n"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vis.metrics(m_metrics)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Fine Tuning a Transfer Learning Model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Retrain the last 3 layers of base model with image augmentation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "for layer in m.layers[:47]:\n",
    "   layer.trainable = False\n",
    "for layer in m.layers[47:]:\n",
    "   layer.trainable = True"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "m.compile(loss='binary_crossentropy', optimizer=\"sgd\", metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "125/125 [==============================] - 50s 399ms/step - loss: 0.1902 - acc: 0.9300 - val_loss: 0.6312 - val_acc: 0.6820\n",
      "Epoch 2/8\n",
      "125/125 [==============================] - 42s 335ms/step - loss: 0.1355 - acc: 0.9440 - val_loss: 0.8368 - val_acc: 0.6783\n",
      "Epoch 3/8\n",
      "125/125 [==============================] - 42s 337ms/step - loss: 0.0909 - acc: 0.9690 - val_loss: 0.6310 - val_acc: 0.6783\n",
      "Epoch 4/8\n",
      "125/125 [==============================] - 42s 337ms/step - loss: 0.0778 - acc: 0.9750 - val_loss: 0.7535 - val_acc: 0.6820\n",
      "Epoch 5/8\n",
      "125/125 [==============================] - 42s 335ms/step - loss: 0.0599 - acc: 0.9790 - val_loss: 1.0659 - val_acc: 0.6801\n",
      "Epoch 6/8\n",
      "125/125 [==============================] - 41s 332ms/step - loss: 0.0751 - acc: 0.9740 - val_loss: 1.1286 - val_acc: 0.6857\n",
      "Epoch 7/8\n",
      "125/125 [==============================] - 42s 334ms/step - loss: 0.0761 - acc: 0.9720 - val_loss: 0.7452 - val_acc: 0.6783\n",
      "Epoch 8/8\n",
      "125/125 [==============================] - 42s 338ms/step - loss: 0.0371 - acc: 0.9910 - val_loss: 0.7481 - val_acc: 0.6691\n"
     ]
    }
   ],
   "source": [
    "m_output_trainable =  m.fit_generator(train_generator, epochs=8, steps_per_epoch=1000 // 8,\n",
    "                            validation_data=val_generator, validation_steps=600 // 8,\n",
    "                            verbose = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.vegalite.v2+json": {
       "$schema": "https://vega.github.io/schema/vega-lite/v2.4.3.json",
       "config": {
        "view": {
         "height": 300,
         "width": 400
        }
       },
       "data": {
        "values": [
         {
          "batch": 1,
          "metrics": "loss",
          "type": "validation",
          "value": 0.631230375784285
         },
         {
          "batch": 2,
          "metrics": "loss",
          "type": "validation",
          "value": 0.8367559686166179
         },
         {
          "batch": 3,
          "metrics": "loss",
          "type": "validation",
          "value": 0.6310281306505203
         },
         {
          "batch": 4,
          "metrics": "loss",
          "type": "validation",
          "value": 0.7534611525205785
         },
         {
          "batch": 5,
          "metrics": "loss",
          "type": "validation",
          "value": 1.0658881804248428
         },
         {
          "batch": 6,
          "metrics": "loss",
          "type": "validation",
          "value": 1.1286484480061678
         },
         {
          "batch": 7,
          "metrics": "loss",
          "type": "validation",
          "value": 0.7452054653207169
         },
         {
          "batch": 8,
          "metrics": "loss",
          "type": "validation",
          "value": 0.7481120452284813
         },
         {
          "batch": 1,
          "metrics": "acc",
          "type": "validation",
          "value": 0.6819852941176471
         },
         {
          "batch": 2,
          "metrics": "acc",
          "type": "validation",
          "value": 0.6783088235294118
         },
         {
          "batch": 3,
          "metrics": "acc",
          "type": "validation",
          "value": 0.6783088235294118
         },
         {
          "batch": 4,
          "metrics": "acc",
          "type": "validation",
          "value": 0.6819852941176471
         },
         {
          "batch": 5,
          "metrics": "acc",
          "type": "validation",
          "value": 0.6801470588235294
         },
         {
          "batch": 6,
          "metrics": "acc",
          "type": "validation",
          "value": 0.6856617647058824
         },
         {
          "batch": 7,
          "metrics": "acc",
          "type": "validation",
          "value": 0.6783088235294118
         },
         {
          "batch": 8,
          "metrics": "acc",
          "type": "validation",
          "value": 0.6691176470588235
         },
         {
          "batch": 1,
          "metrics": "loss",
          "type": "training",
          "value": 0.18545942939060733
         },
         {
          "batch": 2,
          "metrics": "loss",
          "type": "training",
          "value": 0.13020609009083406
         },
         {
          "batch": 3,
          "metrics": "loss",
          "type": "training",
          "value": 0.07943052393859443
         },
         {
          "batch": 4,
          "metrics": "loss",
          "type": "training",
          "value": 0.07867309689448264
         },
         {
          "batch": 5,
          "metrics": "loss",
          "type": "training",
          "value": 0.05754812551062266
         },
         {
          "batch": 6,
          "metrics": "loss",
          "type": "training",
          "value": 0.0696549504773873
         },
         {
          "batch": 7,
          "metrics": "loss",
          "type": "training",
          "value": 0.07564967090493008
         },
         {
          "batch": 8,
          "metrics": "loss",
          "type": "training",
          "value": 0.037414738582494306
         },
         {
          "batch": 1,
          "metrics": "acc",
          "type": "training",
          "value": 0.9298780487804879
         },
         {
          "batch": 2,
          "metrics": "acc",
          "type": "training",
          "value": 0.9463562753036437
         },
         {
          "batch": 3,
          "metrics": "acc",
          "type": "training",
          "value": 0.9714285714285714
         },
         {
          "batch": 4,
          "metrics": "acc",
          "type": "training",
          "value": 0.9746963562753036
         },
         {
          "batch": 5,
          "metrics": "acc",
          "type": "training",
          "value": 0.9796747967479674
         },
         {
          "batch": 6,
          "metrics": "acc",
          "type": "training",
          "value": 0.975609756097561
         },
         {
          "batch": 7,
          "metrics": "acc",
          "type": "training",
          "value": 0.9725609756097561
         },
         {
          "batch": 8,
          "metrics": "acc",
          "type": "training",
          "value": 0.9908906882591093
         }
        ]
       },
       "facet": {
        "column": {
         "field": "metrics",
         "type": "nominal"
        }
       },
       "resolve": {
        "scale": {
         "color": "shared",
         "x": "shared",
         "y": "independent"
        }
       },
       "spec": {
        "layer": [
         {
          "encoding": {
           "color": {
            "field": "type",
            "type": "nominal"
           },
           "tooltip": [
            {
             "field": "batch",
             "type": "quantitative"
            },
            {
             "field": "value",
             "type": "quantitative"
            }
           ],
           "x": {
            "field": "batch",
            "type": "quantitative"
           },
           "y": {
            "field": "value",
            "type": "quantitative"
           }
          },
          "height": 300,
          "mark": {
           "size": 50,
           "type": "circle"
          },
          "width": 300
         },
         {
          "encoding": {
           "color": {
            "field": "type",
            "type": "nominal"
           },
           "x": {
            "field": "batch",
            "type": "quantitative"
           },
           "y": {
            "field": "value",
            "type": "quantitative"
           }
          },
          "height": 300,
          "mark": "line",
          "width": 300
         }
        ]
       }
      },
      "image/png": "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",
      "text/plain": [
       "<VegaLite 2 object>\n",
       "\n",
       "If you see this message, it means the renderer has not been properly enabled\n",
       "for the frontend that you are using. For more information, see\n",
       "https://altair-viz.github.io/user_guide/troubleshooting.html\n"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vis.metrics(m_output_trainable.history)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}