ResNet.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "7aaa9d4b",
   "metadata": {},
   "source": [
    "## Fashion Item Recognition with Residual Neural Network\n",
    "\n",
    "> Antonopoulos Ilias (p3352004) <br />\n",
    "> Ndoja Silva (p3352017) <br />\n",
    "> MSc Data Science AUEB"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d3e6b51d",
   "metadata": {},
   "source": [
    "## Table of Contents\n",
    "\n",
    "- [Data Loading](#Data-Loading)\n",
    "- [Hyperparameter Tuning](#Hyperparameter-Tuning)\n",
    "- [Model Selection](#Model-Selection)\n",
    "- [Evaluation](#Evaluation)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "e88eb046",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-03-08 01:21:04.825219: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:936] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2022-03-08 01:21:04.856152: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:936] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2022-03-08 01:21:04.856612: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:936] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n"
     ]
    }
   ],
   "source": [
    "import gc\n",
    "import itertools\n",
    "\n",
    "import numpy as np\n",
    "import keras_tuner as kt\n",
    "import matplotlib.pyplot as plt\n",
    "import tensorflow as tf\n",
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "from tensorflow.config.experimental import list_physical_devices, set_memory_growth\n",
    "\n",
    "physical_devices = list_physical_devices(\"GPU\")\n",
    "set_memory_growth(physical_devices[0], True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "140836a9",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2.8.0\n"
     ]
    }
   ],
   "source": [
    "print(tf.__version__)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "96edda17",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Num GPUs Available:  1\n"
     ]
    }
   ],
   "source": [
    "print(\"Num GPUs Available: \", len(tf.config.list_physical_devices(\"GPU\")))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "38c99bb4",
   "metadata": {},
   "source": [
    "### Data Loading"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "f57d2b13",
   "metadata": {},
   "outputs": [],
   "source": [
    "fashion_mnist = tf.keras.datasets.fashion_mnist\n",
    "\n",
    "(train_images, train_labels), (test_images, test_labels) = fashion_mnist.load_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "d12d3533",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(60000, 28, 28)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_images.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "3392e806",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([9, 0, 0, ..., 3, 0, 5], dtype=uint8)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "b594e776",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{0, 1, 2, 3, 4, 5, 6, 7, 8, 9}"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "set(train_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "02ce3cc4",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_labels = tf.keras.utils.to_categorical(train_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "d0d3cf3c",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_labels = tf.keras.utils.to_categorical(test_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "d990e8e8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10000, 28, 28)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_images.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0c74d426",
   "metadata": {},
   "source": [
    "This is a dataset of 60,000 28x28 grayscale images of 10 fashion categories,\n",
    "  along with a test set of 10,000 images.\n",
    "  \n",
    "The classes are:\n",
    "  \n",
    "| Label | Description |\n",
    "|:-----:|-------------|\n",
    "|   0   | T-shirt/top |\n",
    "|   1   | Trouser     |\n",
    "|   2   | Pullover    |\n",
    "|   3   | Dress       |\n",
    "|   4   | Coat        |\n",
    "|   5   | Sandal      |\n",
    "|   6   | Shirt       |\n",
    "|   7   | Sneaker     |\n",
    "|   8   | Bag         |\n",
    "|   9   | Ankle boot  |"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "c6b81307",
   "metadata": {},
   "outputs": [],
   "source": [
    "class_names = [\n",
    "    \"T-shirt/top\",\n",
    "    \"Trouser\",\n",
    "    \"Pullover\",\n",
    "    \"Dress\",\n",
    "    \"Coat\",\n",
    "    \"Sandal\",\n",
    "    \"Shirt\",\n",
    "    \"Sneaker\",\n",
    "    \"Bag\",\n",
    "    \"Ankle boot\",\n",
    "]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e78f6372",
   "metadata": {},
   "source": [
    "### Hyperparameter Tuning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "89e40239",
   "metadata": {},
   "outputs": [],
   "source": [
    "SEED = 123456\n",
    "\n",
    "np.random.seed(SEED)\n",
    "tf.random.set_seed(SEED)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "c8b10345",
   "metadata": {},
   "outputs": [],
   "source": [
    "def clean_up(model_):\n",
    "    tf.keras.backend.clear_session()\n",
    "    del model_\n",
    "    gc.collect()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "9bde8bf8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.001"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hparams = kt.HyperParameters()\n",
    "\n",
    "hparams.Fixed(\"optimizer\", value=\"adam\")\n",
    "hparams.Fixed(\"pooling\", value=\"max\")\n",
    "hparams.Choice(\"learning_rate\", values=[1e-3, 1e-4])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "c60e87cb",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-03-08 01:21:05.390402: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:936] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2022-03-08 01:21:05.390811: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:936] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2022-03-08 01:21:05.391177: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:936] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2022-03-08 01:21:05.919644: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:936] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2022-03-08 01:21:05.920022: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:936] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2022-03-08 01:21:05.920407: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:936] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2022-03-08 01:21:05.920749: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1525] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 3069 MB memory:  -> device: 0, name: NVIDIA GeForce GTX 960, pci bus id: 0000:01:00.0, compute capability: 5.2\n"
     ]
    }
   ],
   "source": [
    "tuner = kt.Hyperband(\n",
    "    kt.applications.HyperResNet(input_shape=(28, 28, 1), classes=10),\n",
    "    objective=\"val_accuracy\",\n",
    "    hyperparameters=hparams,\n",
    "    max_epochs=50,  # the maximum number of epochs to train one model\n",
    "    seed=SEED,\n",
    "    directory=\"hparam-tuning\",\n",
    "    project_name=\"resnet\",\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "1619ba71",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Search space summary\n",
      "Default search space size: 6\n",
      "optimizer (Fixed)\n",
      "{'conditions': [], 'value': 'adam'}\n",
      "pooling (Fixed)\n",
      "{'conditions': [], 'value': 'max'}\n",
      "learning_rate (Choice)\n",
      "{'default': 0.001, 'conditions': [], 'values': [0.001, 0.0001], 'ordered': True}\n",
      "version (Choice)\n",
      "{'default': 'v2', 'conditions': [], 'values': ['v1', 'v2', 'next'], 'ordered': False}\n",
      "conv3_depth (Choice)\n",
      "{'default': 4, 'conditions': [], 'values': [4, 8], 'ordered': True}\n",
      "conv4_depth (Choice)\n",
      "{'default': 6, 'conditions': [], 'values': [6, 23, 36], 'ordered': True}\n"
     ]
    }
   ],
   "source": [
    "tuner.search_space_summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "e80f13f9",
   "metadata": {},
   "outputs": [],
   "source": [
    "stop_early = tf.keras.callbacks.EarlyStopping(monitor=\"val_loss\", patience=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "212219f7",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trial 28 Complete [00h 11m 32s]\n",
      "val_accuracy: 0.8125\n",
      "\n",
      "Best val_accuracy So Far: 0.8519999980926514\n",
      "Total elapsed time: 03h 55m 30s\n",
      "INFO:tensorflow:Oracle triggered exit\n"
     ]
    }
   ],
   "source": [
    "tuner.search(\n",
    "    train_images, train_labels, epochs=50, validation_split=0.2, callbacks=[stop_early]\n",
    ")\n",
    "\n",
    "# get the optimal hyperparameters\n",
    "best_hps = tuner.get_best_hyperparameters(num_trials=1)[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "34bafa08",
   "metadata": {},
   "source": [
    "### Model Selection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "f570c8b5",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"ResNet\"\n",
      "__________________________________________________________________________________________________\n",
      " Layer (type)                   Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      " input_1 (InputLayer)           [(None, 28, 28, 1)]  0           []                               \n",
      "                                                                                                  \n",
      " conv1_pad (ZeroPadding2D)      (None, 34, 34, 1)    0           ['input_1[0][0]']                \n",
      "                                                                                                  \n",
      " conv1_conv (Conv2D)            (None, 14, 14, 64)   3200        ['conv1_pad[0][0]']              \n",
      "                                                                                                  \n",
      " pool1_pad (ZeroPadding2D)      (None, 16, 16, 64)   0           ['conv1_conv[0][0]']             \n",
      "                                                                                                  \n",
      " pool1_pool (MaxPooling2D)      (None, 7, 7, 64)     0           ['pool1_pad[0][0]']              \n",
      "                                                                                                  \n",
      " conv2_block1_preact_bn (BatchN  (None, 7, 7, 64)    256         ['pool1_pool[0][0]']             \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " conv2_block1_preact_relu (Acti  (None, 7, 7, 64)    0           ['conv2_block1_preact_bn[0][0]'] \n",
      " vation)                                                                                          \n",
      "                                                                                                  \n",
      " conv2_block1_1_conv (Conv2D)   (None, 7, 7, 64)     4096        ['conv2_block1_preact_relu[0][0]'\n",
      "                                                                 ]                                \n",
      "                                                                                                  \n",
      " conv2_block1_1_bn (BatchNormal  (None, 7, 7, 64)    256         ['conv2_block1_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv2_block1_1_relu (Activatio  (None, 7, 7, 64)    0           ['conv2_block1_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv2_block1_2_pad (ZeroPaddin  (None, 9, 9, 64)    0           ['conv2_block1_1_relu[0][0]']    \n",
      " g2D)                                                                                             \n",
      "                                                                                                  \n",
      " conv2_block1_2_conv (Conv2D)   (None, 7, 7, 64)     36864       ['conv2_block1_2_pad[0][0]']     \n",
      "                                                                                                  \n",
      " conv2_block1_2_bn (BatchNormal  (None, 7, 7, 64)    256         ['conv2_block1_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv2_block1_2_relu (Activatio  (None, 7, 7, 64)    0           ['conv2_block1_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv2_block1_0_conv (Conv2D)   (None, 7, 7, 256)    16640       ['conv2_block1_preact_relu[0][0]'\n",
      "                                                                 ]                                \n",
      "                                                                                                  \n",
      " conv2_block1_3_conv (Conv2D)   (None, 7, 7, 256)    16640       ['conv2_block1_2_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv2_block1_out (Add)         (None, 7, 7, 256)    0           ['conv2_block1_0_conv[0][0]',    \n",
      "                                                                  'conv2_block1_3_conv[0][0]']    \n",
      "                                                                                                  \n",
      " conv2_block2_preact_bn (BatchN  (None, 7, 7, 256)   1024        ['conv2_block1_out[0][0]']       \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " conv2_block2_preact_relu (Acti  (None, 7, 7, 256)   0           ['conv2_block2_preact_bn[0][0]'] \n",
      " vation)                                                                                          \n",
      "                                                                                                  \n",
      " conv2_block2_1_conv (Conv2D)   (None, 7, 7, 64)     16384       ['conv2_block2_preact_relu[0][0]'\n",
      "                                                                 ]                                \n",
      "                                                                                                  \n",
      " conv2_block2_1_bn (BatchNormal  (None, 7, 7, 64)    256         ['conv2_block2_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv2_block2_1_relu (Activatio  (None, 7, 7, 64)    0           ['conv2_block2_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv2_block2_2_pad (ZeroPaddin  (None, 9, 9, 64)    0           ['conv2_block2_1_relu[0][0]']    \n",
      " g2D)                                                                                             \n",
      "                                                                                                  \n",
      " conv2_block2_2_conv (Conv2D)   (None, 7, 7, 64)     36864       ['conv2_block2_2_pad[0][0]']     \n",
      "                                                                                                  \n",
      " conv2_block2_2_bn (BatchNormal  (None, 7, 7, 64)    256         ['conv2_block2_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv2_block2_2_relu (Activatio  (None, 7, 7, 64)    0           ['conv2_block2_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv2_block2_3_conv (Conv2D)   (None, 7, 7, 256)    16640       ['conv2_block2_2_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv2_block2_out (Add)         (None, 7, 7, 256)    0           ['conv2_block1_out[0][0]',       \n",
      "                                                                  'conv2_block2_3_conv[0][0]']    \n",
      "                                                                                                  \n",
      " conv2_block3_preact_bn (BatchN  (None, 7, 7, 256)   1024        ['conv2_block2_out[0][0]']       \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " conv2_block3_preact_relu (Acti  (None, 7, 7, 256)   0           ['conv2_block3_preact_bn[0][0]'] \n",
      " vation)                                                                                          \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                                                                                  \n",
      " conv2_block3_1_conv (Conv2D)   (None, 7, 7, 64)     16384       ['conv2_block3_preact_relu[0][0]'\n",
      "                                                                 ]                                \n",
      "                                                                                                  \n",
      " conv2_block3_1_bn (BatchNormal  (None, 7, 7, 64)    256         ['conv2_block3_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv2_block3_1_relu (Activatio  (None, 7, 7, 64)    0           ['conv2_block3_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv2_block3_2_pad (ZeroPaddin  (None, 9, 9, 64)    0           ['conv2_block3_1_relu[0][0]']    \n",
      " g2D)                                                                                             \n",
      "                                                                                                  \n",
      " conv2_block3_2_conv (Conv2D)   (None, 4, 4, 64)     36864       ['conv2_block3_2_pad[0][0]']     \n",
      "                                                                                                  \n",
      " conv2_block3_2_bn (BatchNormal  (None, 4, 4, 64)    256         ['conv2_block3_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv2_block3_2_relu (Activatio  (None, 4, 4, 64)    0           ['conv2_block3_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " max_pooling2d (MaxPooling2D)   (None, 4, 4, 256)    0           ['conv2_block2_out[0][0]']       \n",
      "                                                                                                  \n",
      " conv2_block3_3_conv (Conv2D)   (None, 4, 4, 256)    16640       ['conv2_block3_2_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv2_block3_out (Add)         (None, 4, 4, 256)    0           ['max_pooling2d[0][0]',          \n",
      "                                                                  'conv2_block3_3_conv[0][0]']    \n",
      "                                                                                                  \n",
      " conv3_block1_preact_bn (BatchN  (None, 4, 4, 256)   1024        ['conv2_block3_out[0][0]']       \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " conv3_block1_preact_relu (Acti  (None, 4, 4, 256)   0           ['conv3_block1_preact_bn[0][0]'] \n",
      " vation)                                                                                          \n",
      "                                                                                                  \n",
      " conv3_block1_1_conv (Conv2D)   (None, 4, 4, 128)    32768       ['conv3_block1_preact_relu[0][0]'\n",
      "                                                                 ]                                \n",
      "                                                                                                  \n",
      " conv3_block1_1_bn (BatchNormal  (None, 4, 4, 128)   512         ['conv3_block1_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv3_block1_1_relu (Activatio  (None, 4, 4, 128)   0           ['conv3_block1_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv3_block1_2_pad (ZeroPaddin  (None, 6, 6, 128)   0           ['conv3_block1_1_relu[0][0]']    \n",
      " g2D)                                                                                             \n",
      "                                                                                                  \n",
      " conv3_block1_2_conv (Conv2D)   (None, 4, 4, 128)    147456      ['conv3_block1_2_pad[0][0]']     \n",
      "                                                                                                  \n",
      " conv3_block1_2_bn (BatchNormal  (None, 4, 4, 128)   512         ['conv3_block1_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv3_block1_2_relu (Activatio  (None, 4, 4, 128)   0           ['conv3_block1_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv3_block1_0_conv (Conv2D)   (None, 4, 4, 512)    131584      ['conv3_block1_preact_relu[0][0]'\n",
      "                                                                 ]                                \n",
      "                                                                                                  \n",
      " conv3_block1_3_conv (Conv2D)   (None, 4, 4, 512)    66048       ['conv3_block1_2_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv3_block1_out (Add)         (None, 4, 4, 512)    0           ['conv3_block1_0_conv[0][0]',    \n",
      "                                                                  'conv3_block1_3_conv[0][0]']    \n",
      "                                                                                                  \n",
      " conv3_block2_preact_bn (BatchN  (None, 4, 4, 512)   2048        ['conv3_block1_out[0][0]']       \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " conv3_block2_preact_relu (Acti  (None, 4, 4, 512)   0           ['conv3_block2_preact_bn[0][0]'] \n",
      " vation)                                                                                          \n",
      "                                                                                                  \n",
      " conv3_block2_1_conv (Conv2D)   (None, 4, 4, 128)    65536       ['conv3_block2_preact_relu[0][0]'\n",
      "                                                                 ]                                \n",
      "                                                                                                  \n",
      " conv3_block2_1_bn (BatchNormal  (None, 4, 4, 128)   512         ['conv3_block2_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv3_block2_1_relu (Activatio  (None, 4, 4, 128)   0           ['conv3_block2_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv3_block2_2_pad (ZeroPaddin  (None, 6, 6, 128)   0           ['conv3_block2_1_relu[0][0]']    \n",
      " g2D)                                                                                             \n",
      "                                                                                                  \n",
      " conv3_block2_2_conv (Conv2D)   (None, 4, 4, 128)    147456      ['conv3_block2_2_pad[0][0]']     \n",
      "                                                                                                  \n",
      " conv3_block2_2_bn (BatchNormal  (None, 4, 4, 128)   512         ['conv3_block2_2_conv[0][0]']    \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv3_block2_2_relu (Activatio  (None, 4, 4, 128)   0           ['conv3_block2_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv3_block2_3_conv (Conv2D)   (None, 4, 4, 512)    66048       ['conv3_block2_2_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv3_block2_out (Add)         (None, 4, 4, 512)    0           ['conv3_block1_out[0][0]',       \n",
      "                                                                  'conv3_block2_3_conv[0][0]']    \n",
      "                                                                                                  \n",
      " conv3_block3_preact_bn (BatchN  (None, 4, 4, 512)   2048        ['conv3_block2_out[0][0]']       \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " conv3_block3_preact_relu (Acti  (None, 4, 4, 512)   0           ['conv3_block3_preact_bn[0][0]'] \n",
      " vation)                                                                                          \n",
      "                                                                                                  \n",
      " conv3_block3_1_conv (Conv2D)   (None, 4, 4, 128)    65536       ['conv3_block3_preact_relu[0][0]'\n",
      "                                                                 ]                                \n",
      "                                                                                                  \n",
      " conv3_block3_1_bn (BatchNormal  (None, 4, 4, 128)   512         ['conv3_block3_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv3_block3_1_relu (Activatio  (None, 4, 4, 128)   0           ['conv3_block3_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv3_block3_2_pad (ZeroPaddin  (None, 6, 6, 128)   0           ['conv3_block3_1_relu[0][0]']    \n",
      " g2D)                                                                                             \n",
      "                                                                                                  \n",
      " conv3_block3_2_conv (Conv2D)   (None, 4, 4, 128)    147456      ['conv3_block3_2_pad[0][0]']     \n",
      "                                                                                                  \n",
      " conv3_block3_2_bn (BatchNormal  (None, 4, 4, 128)   512         ['conv3_block3_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv3_block3_2_relu (Activatio  (None, 4, 4, 128)   0           ['conv3_block3_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv3_block3_3_conv (Conv2D)   (None, 4, 4, 512)    66048       ['conv3_block3_2_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv3_block3_out (Add)         (None, 4, 4, 512)    0           ['conv3_block2_out[0][0]',       \n",
      "                                                                  'conv3_block3_3_conv[0][0]']    \n",
      "                                                                                                  \n",
      " conv3_block4_preact_bn (BatchN  (None, 4, 4, 512)   2048        ['conv3_block3_out[0][0]']       \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " conv3_block4_preact_relu (Acti  (None, 4, 4, 512)   0           ['conv3_block4_preact_bn[0][0]'] \n",
      " vation)                                                                                          \n",
      "                                                                                                  \n",
      " conv3_block4_1_conv (Conv2D)   (None, 4, 4, 128)    65536       ['conv3_block4_preact_relu[0][0]'\n",
      "                                                                 ]                                \n",
      "                                                                                                  \n",
      " conv3_block4_1_bn (BatchNormal  (None, 4, 4, 128)   512         ['conv3_block4_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv3_block4_1_relu (Activatio  (None, 4, 4, 128)   0           ['conv3_block4_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv3_block4_2_pad (ZeroPaddin  (None, 6, 6, 128)   0           ['conv3_block4_1_relu[0][0]']    \n",
      " g2D)                                                                                             \n",
      "                                                                                                  \n",
      " conv3_block4_2_conv (Conv2D)   (None, 2, 2, 128)    147456      ['conv3_block4_2_pad[0][0]']     \n",
      "                                                                                                  \n",
      " conv3_block4_2_bn (BatchNormal  (None, 2, 2, 128)   512         ['conv3_block4_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv3_block4_2_relu (Activatio  (None, 2, 2, 128)   0           ['conv3_block4_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " max_pooling2d_1 (MaxPooling2D)  (None, 2, 2, 512)   0           ['conv3_block3_out[0][0]']       \n",
      "                                                                                                  \n",
      " conv3_block4_3_conv (Conv2D)   (None, 2, 2, 512)    66048       ['conv3_block4_2_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv3_block4_out (Add)         (None, 2, 2, 512)    0           ['max_pooling2d_1[0][0]',        \n",
      "                                                                  'conv3_block4_3_conv[0][0]']    \n",
      "                                                                                                  \n",
      " conv4_block1_preact_bn (BatchN  (None, 2, 2, 512)   2048        ['conv3_block4_out[0][0]']       \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " conv4_block1_preact_relu (Acti  (None, 2, 2, 512)   0           ['conv4_block1_preact_bn[0][0]'] \n",
      " vation)                                                                                          \n",
      "                                                                                                  \n",
      " conv4_block1_1_conv (Conv2D)   (None, 2, 2, 256)    131072      ['conv4_block1_preact_relu[0][0]'\n",
      "                                                                 ]                                \n",
      "                                                                                                  \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " conv4_block1_1_bn (BatchNormal  (None, 2, 2, 256)   1024        ['conv4_block1_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv4_block1_1_relu (Activatio  (None, 2, 2, 256)   0           ['conv4_block1_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv4_block1_2_pad (ZeroPaddin  (None, 4, 4, 256)   0           ['conv4_block1_1_relu[0][0]']    \n",
      " g2D)                                                                                             \n",
      "                                                                                                  \n",
      " conv4_block1_2_conv (Conv2D)   (None, 2, 2, 256)    589824      ['conv4_block1_2_pad[0][0]']     \n",
      "                                                                                                  \n",
      " conv4_block1_2_bn (BatchNormal  (None, 2, 2, 256)   1024        ['conv4_block1_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv4_block1_2_relu (Activatio  (None, 2, 2, 256)   0           ['conv4_block1_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv4_block1_0_conv (Conv2D)   (None, 2, 2, 1024)   525312      ['conv4_block1_preact_relu[0][0]'\n",
      "                                                                 ]                                \n",
      "                                                                                                  \n",
      " conv4_block1_3_conv (Conv2D)   (None, 2, 2, 1024)   263168      ['conv4_block1_2_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv4_block1_out (Add)         (None, 2, 2, 1024)   0           ['conv4_block1_0_conv[0][0]',    \n",
      "                                                                  'conv4_block1_3_conv[0][0]']    \n",
      "                                                                                                  \n",
      " conv4_block2_preact_bn (BatchN  (None, 2, 2, 1024)  4096        ['conv4_block1_out[0][0]']       \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " conv4_block2_preact_relu (Acti  (None, 2, 2, 1024)  0           ['conv4_block2_preact_bn[0][0]'] \n",
      " vation)                                                                                          \n",
      "                                                                                                  \n",
      " conv4_block2_1_conv (Conv2D)   (None, 2, 2, 256)    262144      ['conv4_block2_preact_relu[0][0]'\n",
      "                                                                 ]                                \n",
      "                                                                                                  \n",
      " conv4_block2_1_bn (BatchNormal  (None, 2, 2, 256)   1024        ['conv4_block2_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv4_block2_1_relu (Activatio  (None, 2, 2, 256)   0           ['conv4_block2_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv4_block2_2_pad (ZeroPaddin  (None, 4, 4, 256)   0           ['conv4_block2_1_relu[0][0]']    \n",
      " g2D)                                                                                             \n",
      "                                                                                                  \n",
      " conv4_block2_2_conv (Conv2D)   (None, 2, 2, 256)    589824      ['conv4_block2_2_pad[0][0]']     \n",
      "                                                                                                  \n",
      " conv4_block2_2_bn (BatchNormal  (None, 2, 2, 256)   1024        ['conv4_block2_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv4_block2_2_relu (Activatio  (None, 2, 2, 256)   0           ['conv4_block2_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv4_block2_3_conv (Conv2D)   (None, 2, 2, 1024)   263168      ['conv4_block2_2_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv4_block2_out (Add)         (None, 2, 2, 1024)   0           ['conv4_block1_out[0][0]',       \n",
      "                                                                  'conv4_block2_3_conv[0][0]']    \n",
      "                                                                                                  \n",
      " conv4_block3_preact_bn (BatchN  (None, 2, 2, 1024)  4096        ['conv4_block2_out[0][0]']       \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " conv4_block3_preact_relu (Acti  (None, 2, 2, 1024)  0           ['conv4_block3_preact_bn[0][0]'] \n",
      " vation)                                                                                          \n",
      "                                                                                                  \n",
      " conv4_block3_1_conv (Conv2D)   (None, 2, 2, 256)    262144      ['conv4_block3_preact_relu[0][0]'\n",
      "                                                                 ]                                \n",
      "                                                                                                  \n",
      " conv4_block3_1_bn (BatchNormal  (None, 2, 2, 256)   1024        ['conv4_block3_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv4_block3_1_relu (Activatio  (None, 2, 2, 256)   0           ['conv4_block3_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv4_block3_2_pad (ZeroPaddin  (None, 4, 4, 256)   0           ['conv4_block3_1_relu[0][0]']    \n",
      " g2D)                                                                                             \n",
      "                                                                                                  \n",
      " conv4_block3_2_conv (Conv2D)   (None, 2, 2, 256)    589824      ['conv4_block3_2_pad[0][0]']     \n",
      "                                                                                                  \n",
      " conv4_block3_2_bn (BatchNormal  (None, 2, 2, 256)   1024        ['conv4_block3_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv4_block3_2_relu (Activatio  (None, 2, 2, 256)   0           ['conv4_block3_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv4_block3_3_conv (Conv2D)   (None, 2, 2, 1024)   263168      ['conv4_block3_2_relu[0][0]']    \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                                                                                  \n",
      " conv4_block3_out (Add)         (None, 2, 2, 1024)   0           ['conv4_block2_out[0][0]',       \n",
      "                                                                  'conv4_block3_3_conv[0][0]']    \n",
      "                                                                                                  \n",
      " conv4_block4_preact_bn (BatchN  (None, 2, 2, 1024)  4096        ['conv4_block3_out[0][0]']       \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " conv4_block4_preact_relu (Acti  (None, 2, 2, 1024)  0           ['conv4_block4_preact_bn[0][0]'] \n",
      " vation)                                                                                          \n",
      "                                                                                                  \n",
      " conv4_block4_1_conv (Conv2D)   (None, 2, 2, 256)    262144      ['conv4_block4_preact_relu[0][0]'\n",
      "                                                                 ]                                \n",
      "                                                                                                  \n",
      " conv4_block4_1_bn (BatchNormal  (None, 2, 2, 256)   1024        ['conv4_block4_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv4_block4_1_relu (Activatio  (None, 2, 2, 256)   0           ['conv4_block4_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv4_block4_2_pad (ZeroPaddin  (None, 4, 4, 256)   0           ['conv4_block4_1_relu[0][0]']    \n",
      " g2D)                                                                                             \n",
      "                                                                                                  \n",
      " conv4_block4_2_conv (Conv2D)   (None, 2, 2, 256)    589824      ['conv4_block4_2_pad[0][0]']     \n",
      "                                                                                                  \n",
      " conv4_block4_2_bn (BatchNormal  (None, 2, 2, 256)   1024        ['conv4_block4_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv4_block4_2_relu (Activatio  (None, 2, 2, 256)   0           ['conv4_block4_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv4_block4_3_conv (Conv2D)   (None, 2, 2, 1024)   263168      ['conv4_block4_2_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv4_block4_out (Add)         (None, 2, 2, 1024)   0           ['conv4_block3_out[0][0]',       \n",
      "                                                                  'conv4_block4_3_conv[0][0]']    \n",
      "                                                                                                  \n",
      " conv4_block5_preact_bn (BatchN  (None, 2, 2, 1024)  4096        ['conv4_block4_out[0][0]']       \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " conv4_block5_preact_relu (Acti  (None, 2, 2, 1024)  0           ['conv4_block5_preact_bn[0][0]'] \n",
      " vation)                                                                                          \n",
      "                                                                                                  \n",
      " conv4_block5_1_conv (Conv2D)   (None, 2, 2, 256)    262144      ['conv4_block5_preact_relu[0][0]'\n",
      "                                                                 ]                                \n",
      "                                                                                                  \n",
      " conv4_block5_1_bn (BatchNormal  (None, 2, 2, 256)   1024        ['conv4_block5_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv4_block5_1_relu (Activatio  (None, 2, 2, 256)   0           ['conv4_block5_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv4_block5_2_pad (ZeroPaddin  (None, 4, 4, 256)   0           ['conv4_block5_1_relu[0][0]']    \n",
      " g2D)                                                                                             \n",
      "                                                                                                  \n",
      " conv4_block5_2_conv (Conv2D)   (None, 2, 2, 256)    589824      ['conv4_block5_2_pad[0][0]']     \n",
      "                                                                                                  \n",
      " conv4_block5_2_bn (BatchNormal  (None, 2, 2, 256)   1024        ['conv4_block5_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv4_block5_2_relu (Activatio  (None, 2, 2, 256)   0           ['conv4_block5_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv4_block5_3_conv (Conv2D)   (None, 2, 2, 1024)   263168      ['conv4_block5_2_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv4_block5_out (Add)         (None, 2, 2, 1024)   0           ['conv4_block4_out[0][0]',       \n",
      "                                                                  'conv4_block5_3_conv[0][0]']    \n",
      "                                                                                                  \n",
      " conv4_block6_preact_bn (BatchN  (None, 2, 2, 1024)  4096        ['conv4_block5_out[0][0]']       \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " conv4_block6_preact_relu (Acti  (None, 2, 2, 1024)  0           ['conv4_block6_preact_bn[0][0]'] \n",
      " vation)                                                                                          \n",
      "                                                                                                  \n",
      " conv4_block6_1_conv (Conv2D)   (None, 2, 2, 256)    262144      ['conv4_block6_preact_relu[0][0]'\n",
      "                                                                 ]                                \n",
      "                                                                                                  \n",
      " conv4_block6_1_bn (BatchNormal  (None, 2, 2, 256)   1024        ['conv4_block6_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv4_block6_1_relu (Activatio  (None, 2, 2, 256)   0           ['conv4_block6_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv4_block6_2_pad (ZeroPaddin  (None, 4, 4, 256)   0           ['conv4_block6_1_relu[0][0]']    \n",
      " g2D)                                                                                             \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                                                                                  \n",
      " conv4_block6_2_conv (Conv2D)   (None, 1, 1, 256)    589824      ['conv4_block6_2_pad[0][0]']     \n",
      "                                                                                                  \n",
      " conv4_block6_2_bn (BatchNormal  (None, 1, 1, 256)   1024        ['conv4_block6_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv4_block6_2_relu (Activatio  (None, 1, 1, 256)   0           ['conv4_block6_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " max_pooling2d_2 (MaxPooling2D)  (None, 1, 1, 1024)  0           ['conv4_block5_out[0][0]']       \n",
      "                                                                                                  \n",
      " conv4_block6_3_conv (Conv2D)   (None, 1, 1, 1024)   263168      ['conv4_block6_2_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv4_block6_out (Add)         (None, 1, 1, 1024)   0           ['max_pooling2d_2[0][0]',        \n",
      "                                                                  'conv4_block6_3_conv[0][0]']    \n",
      "                                                                                                  \n",
      " conv5_block1_preact_bn (BatchN  (None, 1, 1, 1024)  4096        ['conv4_block6_out[0][0]']       \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " conv5_block1_preact_relu (Acti  (None, 1, 1, 1024)  0           ['conv5_block1_preact_bn[0][0]'] \n",
      " vation)                                                                                          \n",
      "                                                                                                  \n",
      " conv5_block1_1_conv (Conv2D)   (None, 1, 1, 512)    524288      ['conv5_block1_preact_relu[0][0]'\n",
      "                                                                 ]                                \n",
      "                                                                                                  \n",
      " conv5_block1_1_bn (BatchNormal  (None, 1, 1, 512)   2048        ['conv5_block1_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv5_block1_1_relu (Activatio  (None, 1, 1, 512)   0           ['conv5_block1_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv5_block1_2_pad (ZeroPaddin  (None, 3, 3, 512)   0           ['conv5_block1_1_relu[0][0]']    \n",
      " g2D)                                                                                             \n",
      "                                                                                                  \n",
      " conv5_block1_2_conv (Conv2D)   (None, 1, 1, 512)    2359296     ['conv5_block1_2_pad[0][0]']     \n",
      "                                                                                                  \n",
      " conv5_block1_2_bn (BatchNormal  (None, 1, 1, 512)   2048        ['conv5_block1_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv5_block1_2_relu (Activatio  (None, 1, 1, 512)   0           ['conv5_block1_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv5_block1_0_conv (Conv2D)   (None, 1, 1, 2048)   2099200     ['conv5_block1_preact_relu[0][0]'\n",
      "                                                                 ]                                \n",
      "                                                                                                  \n",
      " conv5_block1_3_conv (Conv2D)   (None, 1, 1, 2048)   1050624     ['conv5_block1_2_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv5_block1_out (Add)         (None, 1, 1, 2048)   0           ['conv5_block1_0_conv[0][0]',    \n",
      "                                                                  'conv5_block1_3_conv[0][0]']    \n",
      "                                                                                                  \n",
      " conv5_block2_preact_bn (BatchN  (None, 1, 1, 2048)  8192        ['conv5_block1_out[0][0]']       \n",
      " ormalization)                                                                                    \n",
      "                                                                                                  \n",
      " conv5_block2_preact_relu (Acti  (None, 1, 1, 2048)  0           ['conv5_block2_preact_bn[0][0]'] \n",
      " vation)                                                                                          \n",
      "                                                                                                  \n",
      " conv5_block2_1_conv (Conv2D)   (None, 1, 1, 512)    1048576     ['conv5_block2_preact_relu[0][0]'\n",
      "                                                                 ]                                \n",
      "                                                                                                  \n",
      " conv5_block2_1_bn (BatchNormal  (None, 1, 1, 512)   2048        ['conv5_block2_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv5_block2_1_relu (Activatio  (None, 1, 1, 512)   0           ['conv5_block2_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv5_block2_2_pad (ZeroPaddin  (None, 3, 3, 512)   0           ['conv5_block2_1_relu[0][0]']    \n",
      " g2D)                                                                                             \n",
      "                                                                                                  \n",
      " conv5_block2_2_conv (Conv2D)   (None, 1, 1, 512)    2359296     ['conv5_block2_2_pad[0][0]']     \n",
      "                                                                                                  \n",
      " conv5_block2_2_bn (BatchNormal  (None, 1, 1, 512)   2048        ['conv5_block2_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv5_block2_2_relu (Activatio  (None, 1, 1, 512)   0           ['conv5_block2_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv5_block2_3_conv (Conv2D)   (None, 1, 1, 2048)   1050624     ['conv5_block2_2_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv5_block2_out (Add)         (None, 1, 1, 2048)   0           ['conv5_block1_out[0][0]',       \n",
      "                                                                  'conv5_block2_3_conv[0][0]']    \n",
      "                                                                                                  \n",
      " conv5_block3_preact_bn (BatchN  (None, 1, 1, 2048)  8192        ['conv5_block2_out[0][0]']       \n",
      " ormalization)                                                                                    \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                                                                                  \n",
      " conv5_block3_preact_relu (Acti  (None, 1, 1, 2048)  0           ['conv5_block3_preact_bn[0][0]'] \n",
      " vation)                                                                                          \n",
      "                                                                                                  \n",
      " conv5_block3_1_conv (Conv2D)   (None, 1, 1, 512)    1048576     ['conv5_block3_preact_relu[0][0]'\n",
      "                                                                 ]                                \n",
      "                                                                                                  \n",
      " conv5_block3_1_bn (BatchNormal  (None, 1, 1, 512)   2048        ['conv5_block3_1_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv5_block3_1_relu (Activatio  (None, 1, 1, 512)   0           ['conv5_block3_1_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv5_block3_2_pad (ZeroPaddin  (None, 3, 3, 512)   0           ['conv5_block3_1_relu[0][0]']    \n",
      " g2D)                                                                                             \n",
      "                                                                                                  \n",
      " conv5_block3_2_conv (Conv2D)   (None, 1, 1, 512)    2359296     ['conv5_block3_2_pad[0][0]']     \n",
      "                                                                                                  \n",
      " conv5_block3_2_bn (BatchNormal  (None, 1, 1, 512)   2048        ['conv5_block3_2_conv[0][0]']    \n",
      " ization)                                                                                         \n",
      "                                                                                                  \n",
      " conv5_block3_2_relu (Activatio  (None, 1, 1, 512)   0           ['conv5_block3_2_bn[0][0]']      \n",
      " n)                                                                                               \n",
      "                                                                                                  \n",
      " conv5_block3_3_conv (Conv2D)   (None, 1, 1, 2048)   1050624     ['conv5_block3_2_relu[0][0]']    \n",
      "                                                                                                  \n",
      " conv5_block3_out (Add)         (None, 1, 1, 2048)   0           ['conv5_block2_out[0][0]',       \n",
      "                                                                  'conv5_block3_3_conv[0][0]']    \n",
      "                                                                                                  \n",
      " post_bn (BatchNormalization)   (None, 1, 1, 2048)   8192        ['conv5_block3_out[0][0]']       \n",
      "                                                                                                  \n",
      " post_relu (Activation)         (None, 1, 1, 2048)   0           ['post_bn[0][0]']                \n",
      "                                                                                                  \n",
      " max_pool (GlobalMaxPooling2D)  (None, 2048)         0           ['post_relu[0][0]']              \n",
      "                                                                                                  \n",
      " probs (Dense)                  (None, 10)           20490       ['max_pool[0][0]']               \n",
      "                                                                                                  \n",
      "==================================================================================================\n",
      "Total params: 23,579,018\n",
      "Trainable params: 23,533,578\n",
      "Non-trainable params: 45,440\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model = tuner.get_best_models(num_models=1)[0]\n",
    "\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "108c1fec",
   "metadata": {},
   "outputs": [],
   "source": [
    "clean_up(model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "a64aae2c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:Detecting that an object or model or tf.train.Checkpoint is being deleted with unrestored values. See the following logs for the specific values in question. To silence these warnings, use `status.expect_partial()`. See https://www.tensorflow.org/api_docs/python/tf/train/Checkpoint#restorefor details about the status object returned by the restore function.\n",
      "WARNING:tensorflow:Value in checkpoint could not be found in the restored object: (root).optimizer.iter\n",
      "WARNING:tensorflow:Value in checkpoint could not be found in the restored object: (root).optimizer.beta_1\n",
      "WARNING:tensorflow:Value in checkpoint could not be found in the restored object: (root).optimizer.beta_2\n",
      "WARNING:tensorflow:Value in checkpoint could not be found in the restored object: (root).optimizer.decay\n",
      "WARNING:tensorflow:Value in checkpoint could not be found in the restored object: (root).optimizer.learning_rate\n",
      "Epoch 1/30\n",
      "1500/1500 [==============================] - 136s 88ms/step - loss: 0.7192 - accuracy: 0.7427 - val_loss: 0.4851 - val_accuracy: 0.8223\n",
      "Epoch 2/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.4418 - accuracy: 0.8407 - val_loss: 0.4034 - val_accuracy: 0.8538\n",
      "Epoch 3/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.3572 - accuracy: 0.8690 - val_loss: 0.3601 - val_accuracy: 0.8688\n",
      "Epoch 4/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.3093 - accuracy: 0.8848 - val_loss: 0.3424 - val_accuracy: 0.8754\n",
      "Epoch 5/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.2773 - accuracy: 0.8975 - val_loss: 0.3628 - val_accuracy: 0.8679\n",
      "Epoch 6/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.2411 - accuracy: 0.9086 - val_loss: 0.3280 - val_accuracy: 0.8812\n",
      "Epoch 7/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.2135 - accuracy: 0.9197 - val_loss: 0.3225 - val_accuracy: 0.8838\n",
      "Epoch 8/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.1905 - accuracy: 0.9298 - val_loss: 0.3213 - val_accuracy: 0.8882\n",
      "Epoch 9/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.1675 - accuracy: 0.9380 - val_loss: 0.3432 - val_accuracy: 0.8854\n",
      "Epoch 10/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.1473 - accuracy: 0.9466 - val_loss: 0.3506 - val_accuracy: 0.8857\n",
      "Epoch 11/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.1296 - accuracy: 0.9517 - val_loss: 0.3752 - val_accuracy: 0.8845\n",
      "Epoch 12/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.1121 - accuracy: 0.9580 - val_loss: 0.3589 - val_accuracy: 0.8883\n",
      "Epoch 13/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.1044 - accuracy: 0.9607 - val_loss: 0.3807 - val_accuracy: 0.8879\n",
      "Epoch 14/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.0937 - accuracy: 0.9655 - val_loss: 0.3894 - val_accuracy: 0.8886\n",
      "Epoch 15/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.0874 - accuracy: 0.9674 - val_loss: 0.3725 - val_accuracy: 0.8932\n",
      "Epoch 16/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.0804 - accuracy: 0.9698 - val_loss: 0.4385 - val_accuracy: 0.8835\n",
      "Epoch 17/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.0736 - accuracy: 0.9736 - val_loss: 0.4088 - val_accuracy: 0.8878\n",
      "Epoch 18/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.0695 - accuracy: 0.9747 - val_loss: 0.3906 - val_accuracy: 0.8982\n",
      "Epoch 19/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.0626 - accuracy: 0.9777 - val_loss: 0.4401 - val_accuracy: 0.8894\n",
      "Epoch 20/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.0625 - accuracy: 0.9777 - val_loss: 0.4220 - val_accuracy: 0.8889\n",
      "Epoch 21/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.0564 - accuracy: 0.9798 - val_loss: 0.4366 - val_accuracy: 0.8909\n",
      "Epoch 22/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.0529 - accuracy: 0.9807 - val_loss: 0.4377 - val_accuracy: 0.8907\n",
      "Epoch 23/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.0534 - accuracy: 0.9815 - val_loss: 0.4670 - val_accuracy: 0.8883\n",
      "Epoch 24/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.0495 - accuracy: 0.9825 - val_loss: 0.5105 - val_accuracy: 0.8777\n",
      "Epoch 25/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.0558 - accuracy: 0.9801 - val_loss: 0.5063 - val_accuracy: 0.8876\n",
      "Epoch 26/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.0420 - accuracy: 0.9855 - val_loss: 0.4605 - val_accuracy: 0.8864\n",
      "Epoch 27/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.0433 - accuracy: 0.9838 - val_loss: 0.4692 - val_accuracy: 0.8935\n",
      "Epoch 28/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.0419 - accuracy: 0.9853 - val_loss: 0.4551 - val_accuracy: 0.8882\n",
      "Epoch 29/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.0389 - accuracy: 0.9863 - val_loss: 0.4644 - val_accuracy: 0.8891\n",
      "Epoch 30/30\n",
      "1500/1500 [==============================] - 131s 87ms/step - loss: 0.0383 - accuracy: 0.9859 - val_loss: 0.4475 - val_accuracy: 0.8982\n",
      "Best epoch: 30\n"
     ]
    }
   ],
   "source": [
    "# build the model with the optimal hyperparameters and train it on the data for 30 epochs\n",
    "model = tuner.hypermodel.build(best_hps)\n",
    "history = model.fit(train_images, train_labels, epochs=30, validation_split=0.2)\n",
    "\n",
    "# keep best epoch\n",
    "val_acc_per_epoch = history.history[\"val_accuracy\"]\n",
    "best_epoch = val_acc_per_epoch.index(max(val_acc_per_epoch)) + 1\n",
    "print(\"Best epoch: %d\" % (best_epoch,))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "8171fcad",
   "metadata": {},
   "outputs": [],
   "source": [
    "clean_up(model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "7a518431",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/30\n",
      "1500/1500 [==============================] - 135s 87ms/step - loss: 0.7165 - accuracy: 0.7435 - val_loss: 0.5116 - val_accuracy: 0.8114\n",
      "Epoch 2/30\n",
      "1500/1500 [==============================] - 130s 87ms/step - loss: 0.4385 - accuracy: 0.8406 - val_loss: 0.4029 - val_accuracy: 0.8529\n",
      "Epoch 3/30\n",
      "1500/1500 [==============================] - 130s 87ms/step - loss: 0.3602 - accuracy: 0.8667 - val_loss: 0.3780 - val_accuracy: 0.8577\n",
      "Epoch 4/30\n",
      "1500/1500 [==============================] - 130s 87ms/step - loss: 0.3098 - accuracy: 0.8849 - val_loss: 0.3392 - val_accuracy: 0.8766\n",
      "Epoch 5/30\n",
      "1500/1500 [==============================] - 130s 87ms/step - loss: 0.2773 - accuracy: 0.8974 - val_loss: 0.3602 - val_accuracy: 0.8668\n",
      "Epoch 6/30\n",
      "1500/1500 [==============================] - 130s 87ms/step - loss: 0.2401 - accuracy: 0.9104 - val_loss: 0.3324 - val_accuracy: 0.8814\n",
      "Epoch 7/30\n",
      "1500/1500 [==============================] - 130s 87ms/step - loss: 0.2173 - accuracy: 0.9184 - val_loss: 0.3207 - val_accuracy: 0.8863\n",
      "Epoch 8/30\n",
      "1500/1500 [==============================] - 130s 87ms/step - loss: 0.1897 - accuracy: 0.9302 - val_loss: 0.3406 - val_accuracy: 0.8840\n",
      "Epoch 9/30\n",
      "1500/1500 [==============================] - 130s 87ms/step - loss: 0.1658 - accuracy: 0.9380 - val_loss: 0.3679 - val_accuracy: 0.8788\n",
      "Epoch 10/30\n",
      "1500/1500 [==============================] - 130s 87ms/step - loss: 0.1443 - accuracy: 0.9457 - val_loss: 0.3545 - val_accuracy: 0.8836\n",
      "Epoch 11/30\n",
      "1500/1500 [==============================] - 130s 87ms/step - loss: 0.1295 - accuracy: 0.9528 - val_loss: 0.3481 - val_accuracy: 0.8919\n",
      "Epoch 12/30\n",
      "1500/1500 [==============================] - 130s 87ms/step - loss: 0.1128 - accuracy: 0.9586 - val_loss: 0.3600 - val_accuracy: 0.8858\n",
      "Epoch 13/30\n",
      "1500/1500 [==============================] - 130s 87ms/step - loss: 0.1040 - accuracy: 0.9614 - val_loss: 0.3713 - val_accuracy: 0.8873\n",
      "Epoch 14/30\n",
      "1500/1500 [==============================] - 130s 87ms/step - loss: 0.0966 - accuracy: 0.9648 - val_loss: 0.3668 - val_accuracy: 0.8895\n",
      "Epoch 15/30\n",
      "1500/1500 [==============================] - 130s 87ms/step - loss: 0.0860 - accuracy: 0.9689 - val_loss: 0.4125 - val_accuracy: 0.8871\n",
      "Epoch 16/30\n",
      "1500/1500 [==============================] - 130s 87ms/step - loss: 0.0769 - accuracy: 0.9727 - val_loss: 0.4356 - val_accuracy: 0.8849\n",
      "Epoch 17/30\n",
      "1500/1500 [==============================] - 130s 87ms/step - loss: 0.0750 - accuracy: 0.9724 - val_loss: 0.3999 - val_accuracy: 0.8947\n",
      "Epoch 18/30\n",
      "1500/1500 [==============================] - 130s 87ms/step - loss: 0.0674 - accuracy: 0.9756 - val_loss: 0.4028 - val_accuracy: 0.8950\n",
      "Epoch 19/30\n",
      "1500/1500 [==============================] - 130s 87ms/step - loss: 0.0651 - accuracy: 0.9762 - val_loss: 0.4005 - val_accuracy: 0.8909\n",
      "Epoch 20/30\n",
      "1500/1500 [==============================] - 130s 87ms/step - loss: 0.0586 - accuracy: 0.9787 - val_loss: 0.4574 - val_accuracy: 0.8903\n",
      "Epoch 21/30\n",
      "1500/1500 [==============================] - 130s 86ms/step - loss: 0.0585 - accuracy: 0.9792 - val_loss: 0.3929 - val_accuracy: 0.8971\n",
      "Epoch 22/30\n",
      "1500/1500 [==============================] - 130s 86ms/step - loss: 0.0536 - accuracy: 0.9811 - val_loss: 0.4208 - val_accuracy: 0.8934\n",
      "Epoch 23/30\n",
      "1500/1500 [==============================] - 130s 87ms/step - loss: 0.0533 - accuracy: 0.9808 - val_loss: 0.4283 - val_accuracy: 0.8907\n",
      "Epoch 24/30\n",
      "1500/1500 [==============================] - 130s 86ms/step - loss: 0.0491 - accuracy: 0.9825 - val_loss: 0.4450 - val_accuracy: 0.8889\n",
      "Epoch 25/30\n",
      "1500/1500 [==============================] - 130s 87ms/step - loss: 0.0466 - accuracy: 0.9839 - val_loss: 0.4029 - val_accuracy: 0.8956\n",
      "Epoch 26/30\n",
      "1500/1500 [==============================] - 130s 87ms/step - loss: 0.0459 - accuracy: 0.9837 - val_loss: 0.4566 - val_accuracy: 0.8889\n",
      "Epoch 27/30\n",
      "1500/1500 [==============================] - 130s 87ms/step - loss: 0.0428 - accuracy: 0.9846 - val_loss: 0.4467 - val_accuracy: 0.8939\n",
      "Epoch 28/30\n",
      "1500/1500 [==============================] - 130s 87ms/step - loss: 0.0397 - accuracy: 0.9858 - val_loss: 0.4632 - val_accuracy: 0.8916\n",
      "Epoch 29/30\n",
      "1500/1500 [==============================] - 130s 87ms/step - loss: 0.0377 - accuracy: 0.9864 - val_loss: 0.4577 - val_accuracy: 0.8936\n",
      "Epoch 30/30\n",
      "1500/1500 [==============================] - 130s 87ms/step - loss: 0.0371 - accuracy: 0.9867 - val_loss: 0.4467 - val_accuracy: 0.8933\n"
     ]
    }
   ],
   "source": [
    "hypermodel = tuner.hypermodel.build(best_hps)\n",
    "\n",
    "# retrain the model\n",
    "history = hypermodel.fit(\n",
    "    train_images, train_labels, epochs=best_epoch, validation_split=0.2\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c234c695",
   "metadata": {},
   "source": [
    "### Evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "92200b59",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[test loss, test accuracy]: [0.45618003606796265, 0.894599974155426]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-03-08 07:27:27.833667: W tensorflow/core/common_runtime/bfc_allocator.cc:275] Allocator (GPU_0_bfc) ran out of memory trying to allocate 2.29GiB with freed_by_count=0. The caller indicates that this is not a failure, but may mean that there could be performance gains if more memory were available.\n"
     ]
    }
   ],
   "source": [
    "eval_result = hypermodel.evaluate(test_images, test_labels, verbose=3)\n",
    "print(\"[test loss, test accuracy]:\", eval_result)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "5e268801",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_history(hs, epochs, metric):\n",
    "    print()\n",
    "    plt.style.use(\"dark_background\")\n",
    "    plt.rcParams[\"figure.figsize\"] = [15, 8]\n",
    "    plt.rcParams[\"font.size\"] = 16\n",
    "    plt.clf()\n",
    "    for label in hs:\n",
    "        plt.plot(\n",
    "            hs[label].history[metric],\n",
    "            label=\"{0:s} train {1:s}\".format(label, metric),\n",
    "            linewidth=2,\n",
    "        )\n",
    "        plt.plot(\n",
    "            hs[label].history[\"val_{0:s}\".format(metric)],\n",
    "            label=\"{0:s} validation {1:s}\".format(label, metric),\n",
    "            linewidth=2,\n",
    "        )\n",
    "    x_ticks = np.arange(0, epochs + 1, epochs / 10)\n",
    "    x_ticks[0] += 1\n",
    "    plt.xticks(x_ticks)\n",
    "    plt.ylim((0, 1))\n",
    "    plt.xlabel(\"Epochs\")\n",
    "    plt.ylabel(\"Loss\" if metric == \"loss\" else \"Accuracy\")\n",
    "    plt.legend()\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "17c36b94",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Loss          : 0.03711\n",
      "Validation Loss     : 0.44671\n",
      "Test Loss           : 0.45618\n",
      "-------------------\n",
      "Train Accuracy      : 0.98671\n",
      "Validation Accuracy : 0.89333\n",
      "Test Accuracy       : 0.89460\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(\"Train Loss          : {0:.5f}\".format(history.history[\"loss\"][-1]))\n",
    "print(\"Validation Loss     : {0:.5f}\".format(history.history[\"val_loss\"][-1]))\n",
    "print(\"Test Loss           : {0:.5f}\".format(eval_result[0]))\n",
    "print(\"-------------------\")\n",
    "print(\"Train Accuracy      : {0:.5f}\".format(history.history[\"accuracy\"][-1]))\n",
    "print(\"Validation Accuracy : {0:.5f}\".format(history.history[\"val_accuracy\"][-1]))\n",
    "print(\"Test Accuracy       : {0:.5f}\".format(eval_result[1]))\n",
    "\n",
    "# Plot train and validation error per epoch.\n",
    "plot_history(hs={\"ResNet\": history}, epochs=best_epoch, metric=\"loss\")\n",
    "plot_history(hs={\"ResNet\": history}, epochs=best_epoch, metric=\"accuracy\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "cf6290d6",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_confusion_matrix(\n",
    "    cm, classes, normalize=False, title=\"Confusion matrix\", cmap=plt.cm.PuBuGn\n",
    "):\n",
    "\n",
    "    plt.style.use(\"default\")\n",
    "    plt.rcParams[\"figure.figsize\"] = [11, 9]\n",
    "    plt.imshow(cm, interpolation=\"nearest\", cmap=cmap)\n",
    "    plt.title(title)\n",
    "    plt.colorbar()\n",
    "    tick_marks = np.arange(len(classes))\n",
    "    plt.xticks(tick_marks, classes, rotation=90)\n",
    "    plt.yticks(tick_marks, classes)\n",
    "\n",
    "    if normalize:\n",
    "        cm = cm.astype(\"float\") / cm.sum(axis=1)[:, np.newaxis]\n",
    "\n",
    "    thresh = cm.max() / 2.0\n",
    "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
    "        plt.text(\n",
    "            j,\n",
    "            i,\n",
    "            cm[i, j],\n",
    "            horizontalalignment=\"center\",\n",
    "            color=\"white\" if cm[i, j] > thresh else \"black\",\n",
    "        )\n",
    "\n",
    "    plt.tight_layout()\n",
    "    plt.ylabel(\"True label\")\n",
    "    plt.xlabel(\"Predicted label\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "446e56fa",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1100x900 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "(train_images, train_labels), (test_images, test_labels) = fashion_mnist.load_data()\n",
    "\n",
    "# Predict the values from the validation dataset\n",
    "Y_pred = hypermodel.predict(test_images)\n",
    "\n",
    "# Convert predictions classes to one hot vectors\n",
    "Y_pred_classes = np.argmax(Y_pred, axis=1)\n",
    "\n",
    "# compute the confusion matrix\n",
    "confusion_mtx = confusion_matrix(test_labels, Y_pred_classes)\n",
    "\n",
    "# plot the confusion matrix\n",
    "plot_confusion_matrix(\n",
    "    confusion_mtx,\n",
    "    classes=class_names,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "703bec9e",
   "metadata": {},
   "outputs": [],
   "source": [
    "incorrect = []\n",
    "for i in range(len(test_labels)):\n",
    "    if not Y_pred_classes[i] == test_labels[i]:\n",
    "        incorrect.append(i)\n",
    "    if len(incorrect) == 4:\n",
    "        break"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "df830269",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Predicted Label : Coat\\nActual Label : Shirt')"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x1000 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(2, 2, figsize=(12, 6))\n",
    "fig.set_size_inches(10, 10)\n",
    "ax[0, 0].imshow(test_images[incorrect[0]].reshape(28, 28), cmap=\"gray\")\n",
    "\n",
    "ax[0, 0].set_title(\n",
    "    \"Predicted Label : \"\n",
    "    + class_names[Y_pred_classes[incorrect[0]]]\n",
    "    + \"\\n\"\n",
    "    + \"Actual Label : \"\n",
    "    + class_names[test_labels[incorrect[0]]]\n",
    ")\n",
    "ax[0, 1].imshow(test_images[incorrect[1]].reshape(28, 28), cmap=\"gray\")\n",
    "ax[0, 1].set_title(\n",
    "    \"Predicted Label : \"\n",
    "    + class_names[Y_pred_classes[incorrect[1]]]\n",
    "    + \"\\n\"\n",
    "    + \"Actual Label : \"\n",
    "    + class_names[test_labels[incorrect[1]]]\n",
    ")\n",
    "ax[1, 0].imshow(test_images[incorrect[2]].reshape(28, 28), cmap=\"gray\")\n",
    "ax[1, 0].set_title(\n",
    "    \"Predicted Label : \"\n",
    "    + class_names[Y_pred_classes[incorrect[2]]]\n",
    "    + \"\\n\"\n",
    "    + \"Actual Label : \"\n",
    "    + class_names[test_labels[incorrect[2]]]\n",
    ")\n",
    "ax[1, 1].imshow(test_images[incorrect[3]].reshape(28, 28), cmap=\"gray\")\n",
    "ax[1, 1].set_title(\n",
    "    \"Predicted Label : \"\n",
    "    + class_names[Y_pred_classes[incorrect[3]]]\n",
    "    + \"\\n\"\n",
    "    + \"Actual Label : \"\n",
    "    + class_names[test_labels[incorrect[3]]]\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": "3.10.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}