Going_Deeper/ResNet.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# ResNet"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "import tensorflow_datasets as tfds"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plain Networks"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Conv block"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def conv_batch(layer, channel, kernel_size, padding='same', strides=1, activation='relu', name='0'):\n",
    "    x = keras.layers.Conv2D(filters=channel,\n",
    "                            kernel_size=kernel_size,\n",
    "                            kernel_initializer='he_normal',\n",
    "                            kernel_regularizer=keras.regularizers.l2(1e-4),\n",
    "                            padding=padding,\n",
    "                            strides=strides,\n",
    "                            name=f'conv_{name}')(layer)\n",
    "    x = keras.layers.BatchNormalization(name=f'batch_{name}')(x)\n",
    "    if activation:\n",
    "        x = keras.layers.Activation(activation, name=f'act_{name}')(x)\n",
    "    \n",
    "    return x"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plain block "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def build_plain_block(input_layer, block_num=0, num_cnn=3, channel=64, is_50=True):\n",
    "    x = input_layer\n",
    "    \n",
    "    # plain_50\n",
    "    if is_50:\n",
    "        for i in range(num_cnn):\n",
    "            if block_num > 0 and i==0:\n",
    "                x = conv_batch(x, channel, (1,1), strides=2, name=f'{block_num}_{i}1')\n",
    "                x = conv_batch(x, channel, (3,3), name=f'{block_num}_{i}2')\n",
    "                x = conv_batch(x, channel*4, (1,1), name=f'{block_num}_{i}3')\n",
    "            else:\n",
    "                x = conv_batch(x, channel, (1,1), name=f'{block_num}_{i}1')\n",
    "                x = conv_batch(x, channel, (3,3), name=f'{block_num}_{i}2')\n",
    "                x = conv_batch(x, channel*4, (1,1), name=f'{block_num}_{i}3')\n",
    "    # plain_34\n",
    "    else:\n",
    "        for i in range(num_cnn):\n",
    "            if block_num > 0 and i==0:\n",
    "                x = conv_batch(x, channel, (3,3), strides=2, name=f'{block_num}_{i}1')\n",
    "                x = conv_batch(x, channel, (3,3), name=f'{block_num}_{i}2')\n",
    "            else:\n",
    "                x = conv_batch(x, channel, (3,3), name=f'{block_num}_{i}1')\n",
    "                x = conv_batch(x, channel, (3,3), name=f'{block_num}_{i}2')\n",
    "\n",
    "    return x"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###  Build plain network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    " def build_plain(num_cnn_list, channel_list, is_50=True, cats_vs_dogs=False, input_shape=(32,32,3), num_classes=10, name='Plain_50'):\n",
    "    \n",
    "    assert len(num_cnn_list) == len(channel_list)\n",
    "    \n",
    "    input_layer = keras.layers.Input(shape=input_shape, name='Input')\n",
    "    \n",
    "    # first layer\n",
    "    conv1 = conv_batch(input_layer, 64, (7,7), strides=2, name='first')\n",
    "    plain = keras.layers.MaxPool2D(pool_size=(2,2), strides=2, name='maxpool_first')(conv1)\n",
    "    \n",
    "    # plain block\n",
    "    for block_num, (cnn, channel) in enumerate(zip(num_cnn_list, channel_list)):\n",
    "        plain = build_plain_block(plain, block_num, cnn, channel, is_50)\n",
    "    \n",
    "    gap = keras.layers.GlobalAveragePooling2D(name='GAP')(plain)\n",
    "    \n",
    "    if cats_vs_dogs:\n",
    "        output = keras.layers.Dense(1, activation='sigmoid',kernel_initializer='he_normal', name='FC')(gap)\n",
    "    else:\n",
    "        output = keras.layers.Dense(num_classes, activation='softmax',kernel_initializer='he_normal', name='FC')(gap)\n",
    "    \n",
    "    model = keras.Model(inputs=input_layer, outputs=output, name=name)\n",
    "    \n",
    "    return model\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plain-34"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"Plain_34\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "Input (InputLayer)           [(None, 224, 224, 3)]     0         \n",
      "_________________________________________________________________\n",
      "conv_first (Conv2D)          (None, 112, 112, 64)      9472      \n",
      "_________________________________________________________________\n",
      "batch_first (BatchNormalizat (None, 112, 112, 64)      256       \n",
      "_________________________________________________________________\n",
      "act_first (Activation)       (None, 112, 112, 64)      0         \n",
      "_________________________________________________________________\n",
      "maxpool_first (MaxPooling2D) (None, 56, 56, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv_0_01 (Conv2D)           (None, 56, 56, 64)        36928     \n",
      "_________________________________________________________________\n",
      "batch_0_01 (BatchNormalizati (None, 56, 56, 64)        256       \n",
      "_________________________________________________________________\n",
      "act_0_01 (Activation)        (None, 56, 56, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv_0_02 (Conv2D)           (None, 56, 56, 64)        36928     \n",
      "_________________________________________________________________\n",
      "batch_0_02 (BatchNormalizati (None, 56, 56, 64)        256       \n",
      "_________________________________________________________________\n",
      "act_0_02 (Activation)        (None, 56, 56, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv_0_11 (Conv2D)           (None, 56, 56, 64)        36928     \n",
      "_________________________________________________________________\n",
      "batch_0_11 (BatchNormalizati (None, 56, 56, 64)        256       \n",
      "_________________________________________________________________\n",
      "act_0_11 (Activation)        (None, 56, 56, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv_0_12 (Conv2D)           (None, 56, 56, 64)        36928     \n",
      "_________________________________________________________________\n",
      "batch_0_12 (BatchNormalizati (None, 56, 56, 64)        256       \n",
      "_________________________________________________________________\n",
      "act_0_12 (Activation)        (None, 56, 56, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv_0_21 (Conv2D)           (None, 56, 56, 64)        36928     \n",
      "_________________________________________________________________\n",
      "batch_0_21 (BatchNormalizati (None, 56, 56, 64)        256       \n",
      "_________________________________________________________________\n",
      "act_0_21 (Activation)        (None, 56, 56, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv_0_22 (Conv2D)           (None, 56, 56, 64)        36928     \n",
      "_________________________________________________________________\n",
      "batch_0_22 (BatchNormalizati (None, 56, 56, 64)        256       \n",
      "_________________________________________________________________\n",
      "act_0_22 (Activation)        (None, 56, 56, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv_1_01 (Conv2D)           (None, 28, 28, 128)       73856     \n",
      "_________________________________________________________________\n",
      "batch_1_01 (BatchNormalizati (None, 28, 28, 128)       512       \n",
      "_________________________________________________________________\n",
      "act_1_01 (Activation)        (None, 28, 28, 128)       0         \n",
      "_________________________________________________________________\n",
      "conv_1_02 (Conv2D)           (None, 28, 28, 128)       147584    \n",
      "_________________________________________________________________\n",
      "batch_1_02 (BatchNormalizati (None, 28, 28, 128)       512       \n",
      "_________________________________________________________________\n",
      "act_1_02 (Activation)        (None, 28, 28, 128)       0         \n",
      "_________________________________________________________________\n",
      "conv_1_11 (Conv2D)           (None, 28, 28, 128)       147584    \n",
      "_________________________________________________________________\n",
      "batch_1_11 (BatchNormalizati (None, 28, 28, 128)       512       \n",
      "_________________________________________________________________\n",
      "act_1_11 (Activation)        (None, 28, 28, 128)       0         \n",
      "_________________________________________________________________\n",
      "conv_1_12 (Conv2D)           (None, 28, 28, 128)       147584    \n",
      "_________________________________________________________________\n",
      "batch_1_12 (BatchNormalizati (None, 28, 28, 128)       512       \n",
      "_________________________________________________________________\n",
      "act_1_12 (Activation)        (None, 28, 28, 128)       0         \n",
      "_________________________________________________________________\n",
      "conv_1_21 (Conv2D)           (None, 28, 28, 128)       147584    \n",
      "_________________________________________________________________\n",
      "batch_1_21 (BatchNormalizati (None, 28, 28, 128)       512       \n",
      "_________________________________________________________________\n",
      "act_1_21 (Activation)        (None, 28, 28, 128)       0         \n",
      "_________________________________________________________________\n",
      "conv_1_22 (Conv2D)           (None, 28, 28, 128)       147584    \n",
      "_________________________________________________________________\n",
      "batch_1_22 (BatchNormalizati (None, 28, 28, 128)       512       \n",
      "_________________________________________________________________\n",
      "act_1_22 (Activation)        (None, 28, 28, 128)       0         \n",
      "_________________________________________________________________\n",
      "conv_1_31 (Conv2D)           (None, 28, 28, 128)       147584    \n",
      "_________________________________________________________________\n",
      "batch_1_31 (BatchNormalizati (None, 28, 28, 128)       512       \n",
      "_________________________________________________________________\n",
      "act_1_31 (Activation)        (None, 28, 28, 128)       0         \n",
      "_________________________________________________________________\n",
      "conv_1_32 (Conv2D)           (None, 28, 28, 128)       147584    \n",
      "_________________________________________________________________\n",
      "batch_1_32 (BatchNormalizati (None, 28, 28, 128)       512       \n",
      "_________________________________________________________________\n",
      "act_1_32 (Activation)        (None, 28, 28, 128)       0         \n",
      "_________________________________________________________________\n",
      "conv_2_01 (Conv2D)           (None, 14, 14, 256)       295168    \n",
      "_________________________________________________________________\n",
      "batch_2_01 (BatchNormalizati (None, 14, 14, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_2_01 (Activation)        (None, 14, 14, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_2_02 (Conv2D)           (None, 14, 14, 256)       590080    \n",
      "_________________________________________________________________\n",
      "batch_2_02 (BatchNormalizati (None, 14, 14, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_2_02 (Activation)        (None, 14, 14, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_2_11 (Conv2D)           (None, 14, 14, 256)       590080    \n",
      "_________________________________________________________________\n",
      "batch_2_11 (BatchNormalizati (None, 14, 14, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_2_11 (Activation)        (None, 14, 14, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_2_12 (Conv2D)           (None, 14, 14, 256)       590080    \n",
      "_________________________________________________________________\n",
      "batch_2_12 (BatchNormalizati (None, 14, 14, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_2_12 (Activation)        (None, 14, 14, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_2_21 (Conv2D)           (None, 14, 14, 256)       590080    \n",
      "_________________________________________________________________\n",
      "batch_2_21 (BatchNormalizati (None, 14, 14, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_2_21 (Activation)        (None, 14, 14, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_2_22 (Conv2D)           (None, 14, 14, 256)       590080    \n",
      "_________________________________________________________________\n",
      "batch_2_22 (BatchNormalizati (None, 14, 14, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_2_22 (Activation)        (None, 14, 14, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_2_31 (Conv2D)           (None, 14, 14, 256)       590080    \n",
      "_________________________________________________________________\n",
      "batch_2_31 (BatchNormalizati (None, 14, 14, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_2_31 (Activation)        (None, 14, 14, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_2_32 (Conv2D)           (None, 14, 14, 256)       590080    \n",
      "_________________________________________________________________\n",
      "batch_2_32 (BatchNormalizati (None, 14, 14, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_2_32 (Activation)        (None, 14, 14, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_2_41 (Conv2D)           (None, 14, 14, 256)       590080    \n",
      "_________________________________________________________________\n",
      "batch_2_41 (BatchNormalizati (None, 14, 14, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_2_41 (Activation)        (None, 14, 14, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_2_42 (Conv2D)           (None, 14, 14, 256)       590080    \n",
      "_________________________________________________________________\n",
      "batch_2_42 (BatchNormalizati (None, 14, 14, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_2_42 (Activation)        (None, 14, 14, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_2_51 (Conv2D)           (None, 14, 14, 256)       590080    \n",
      "_________________________________________________________________\n",
      "batch_2_51 (BatchNormalizati (None, 14, 14, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_2_51 (Activation)        (None, 14, 14, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_2_52 (Conv2D)           (None, 14, 14, 256)       590080    \n",
      "_________________________________________________________________\n",
      "batch_2_52 (BatchNormalizati (None, 14, 14, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_2_52 (Activation)        (None, 14, 14, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_3_01 (Conv2D)           (None, 7, 7, 512)         1180160   \n",
      "_________________________________________________________________\n",
      "batch_3_01 (BatchNormalizati (None, 7, 7, 512)         2048      \n",
      "_________________________________________________________________\n",
      "act_3_01 (Activation)        (None, 7, 7, 512)         0         \n",
      "_________________________________________________________________\n",
      "conv_3_02 (Conv2D)           (None, 7, 7, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "batch_3_02 (BatchNormalizati (None, 7, 7, 512)         2048      \n",
      "_________________________________________________________________\n",
      "act_3_02 (Activation)        (None, 7, 7, 512)         0         \n",
      "_________________________________________________________________\n",
      "conv_3_11 (Conv2D)           (None, 7, 7, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "batch_3_11 (BatchNormalizati (None, 7, 7, 512)         2048      \n",
      "_________________________________________________________________\n",
      "act_3_11 (Activation)        (None, 7, 7, 512)         0         \n",
      "_________________________________________________________________\n",
      "conv_3_12 (Conv2D)           (None, 7, 7, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "batch_3_12 (BatchNormalizati (None, 7, 7, 512)         2048      \n",
      "_________________________________________________________________\n",
      "act_3_12 (Activation)        (None, 7, 7, 512)         0         \n",
      "_________________________________________________________________\n",
      "conv_3_21 (Conv2D)           (None, 7, 7, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "batch_3_21 (BatchNormalizati (None, 7, 7, 512)         2048      \n",
      "_________________________________________________________________\n",
      "act_3_21 (Activation)        (None, 7, 7, 512)         0         \n",
      "_________________________________________________________________\n",
      "conv_3_22 (Conv2D)           (None, 7, 7, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "batch_3_22 (BatchNormalizati (None, 7, 7, 512)         2048      \n",
      "_________________________________________________________________\n",
      "act_3_22 (Activation)        (None, 7, 7, 512)         0         \n",
      "_________________________________________________________________\n",
      "GAP (GlobalAveragePooling2D) (None, 512)               0         \n",
      "_________________________________________________________________\n",
      "FC (Dense)                   (None, 10)                5130      \n",
      "=================================================================\n",
      "Total params: 21,138,826\n",
      "Trainable params: 21,123,594\n",
      "Non-trainable params: 15,232\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "plain_34 = build_plain(num_cnn_list=[3, 4, 6, 3],\n",
    "                       channel_list=[64, 128, 256, 512],\n",
    "                       is_50=False,\n",
    "                       input_shape=(224,224,3),\n",
    "                       name='Plain_34')\n",
    "plain_34.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plain-50"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"Plain_50\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "Input (InputLayer)           [(None, 224, 224, 3)]     0         \n",
      "_________________________________________________________________\n",
      "conv_first (Conv2D)          (None, 112, 112, 64)      9472      \n",
      "_________________________________________________________________\n",
      "batch_first (BatchNormalizat (None, 112, 112, 64)      256       \n",
      "_________________________________________________________________\n",
      "act_first (Activation)       (None, 112, 112, 64)      0         \n",
      "_________________________________________________________________\n",
      "maxpool_first (MaxPooling2D) (None, 56, 56, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv_0_01 (Conv2D)           (None, 56, 56, 64)        4160      \n",
      "_________________________________________________________________\n",
      "batch_0_01 (BatchNormalizati (None, 56, 56, 64)        256       \n",
      "_________________________________________________________________\n",
      "act_0_01 (Activation)        (None, 56, 56, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv_0_02 (Conv2D)           (None, 56, 56, 64)        36928     \n",
      "_________________________________________________________________\n",
      "batch_0_02 (BatchNormalizati (None, 56, 56, 64)        256       \n",
      "_________________________________________________________________\n",
      "act_0_02 (Activation)        (None, 56, 56, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv_0_03 (Conv2D)           (None, 56, 56, 256)       16640     \n",
      "_________________________________________________________________\n",
      "batch_0_03 (BatchNormalizati (None, 56, 56, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_0_03 (Activation)        (None, 56, 56, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_0_11 (Conv2D)           (None, 56, 56, 64)        16448     \n",
      "_________________________________________________________________\n",
      "batch_0_11 (BatchNormalizati (None, 56, 56, 64)        256       \n",
      "_________________________________________________________________\n",
      "act_0_11 (Activation)        (None, 56, 56, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv_0_12 (Conv2D)           (None, 56, 56, 64)        36928     \n",
      "_________________________________________________________________\n",
      "batch_0_12 (BatchNormalizati (None, 56, 56, 64)        256       \n",
      "_________________________________________________________________\n",
      "act_0_12 (Activation)        (None, 56, 56, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv_0_13 (Conv2D)           (None, 56, 56, 256)       16640     \n",
      "_________________________________________________________________\n",
      "batch_0_13 (BatchNormalizati (None, 56, 56, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_0_13 (Activation)        (None, 56, 56, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_0_21 (Conv2D)           (None, 56, 56, 64)        16448     \n",
      "_________________________________________________________________\n",
      "batch_0_21 (BatchNormalizati (None, 56, 56, 64)        256       \n",
      "_________________________________________________________________\n",
      "act_0_21 (Activation)        (None, 56, 56, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv_0_22 (Conv2D)           (None, 56, 56, 64)        36928     \n",
      "_________________________________________________________________\n",
      "batch_0_22 (BatchNormalizati (None, 56, 56, 64)        256       \n",
      "_________________________________________________________________\n",
      "act_0_22 (Activation)        (None, 56, 56, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv_0_23 (Conv2D)           (None, 56, 56, 256)       16640     \n",
      "_________________________________________________________________\n",
      "batch_0_23 (BatchNormalizati (None, 56, 56, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_0_23 (Activation)        (None, 56, 56, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_1_01 (Conv2D)           (None, 28, 28, 128)       32896     \n",
      "_________________________________________________________________\n",
      "batch_1_01 (BatchNormalizati (None, 28, 28, 128)       512       \n",
      "_________________________________________________________________\n",
      "act_1_01 (Activation)        (None, 28, 28, 128)       0         \n",
      "_________________________________________________________________\n",
      "conv_1_02 (Conv2D)           (None, 28, 28, 128)       147584    \n",
      "_________________________________________________________________\n",
      "batch_1_02 (BatchNormalizati (None, 28, 28, 128)       512       \n",
      "_________________________________________________________________\n",
      "act_1_02 (Activation)        (None, 28, 28, 128)       0         \n",
      "_________________________________________________________________\n",
      "conv_1_03 (Conv2D)           (None, 28, 28, 512)       66048     \n",
      "_________________________________________________________________\n",
      "batch_1_03 (BatchNormalizati (None, 28, 28, 512)       2048      \n",
      "_________________________________________________________________\n",
      "act_1_03 (Activation)        (None, 28, 28, 512)       0         \n",
      "_________________________________________________________________\n",
      "conv_1_11 (Conv2D)           (None, 28, 28, 128)       65664     \n",
      "_________________________________________________________________\n",
      "batch_1_11 (BatchNormalizati (None, 28, 28, 128)       512       \n",
      "_________________________________________________________________\n",
      "act_1_11 (Activation)        (None, 28, 28, 128)       0         \n",
      "_________________________________________________________________\n",
      "conv_1_12 (Conv2D)           (None, 28, 28, 128)       147584    \n",
      "_________________________________________________________________\n",
      "batch_1_12 (BatchNormalizati (None, 28, 28, 128)       512       \n",
      "_________________________________________________________________\n",
      "act_1_12 (Activation)        (None, 28, 28, 128)       0         \n",
      "_________________________________________________________________\n",
      "conv_1_13 (Conv2D)           (None, 28, 28, 512)       66048     \n",
      "_________________________________________________________________\n",
      "batch_1_13 (BatchNormalizati (None, 28, 28, 512)       2048      \n",
      "_________________________________________________________________\n",
      "act_1_13 (Activation)        (None, 28, 28, 512)       0         \n",
      "_________________________________________________________________\n",
      "conv_1_21 (Conv2D)           (None, 28, 28, 128)       65664     \n",
      "_________________________________________________________________\n",
      "batch_1_21 (BatchNormalizati (None, 28, 28, 128)       512       \n",
      "_________________________________________________________________\n",
      "act_1_21 (Activation)        (None, 28, 28, 128)       0         \n",
      "_________________________________________________________________\n",
      "conv_1_22 (Conv2D)           (None, 28, 28, 128)       147584    \n",
      "_________________________________________________________________\n",
      "batch_1_22 (BatchNormalizati (None, 28, 28, 128)       512       \n",
      "_________________________________________________________________\n",
      "act_1_22 (Activation)        (None, 28, 28, 128)       0         \n",
      "_________________________________________________________________\n",
      "conv_1_23 (Conv2D)           (None, 28, 28, 512)       66048     \n",
      "_________________________________________________________________\n",
      "batch_1_23 (BatchNormalizati (None, 28, 28, 512)       2048      \n",
      "_________________________________________________________________\n",
      "act_1_23 (Activation)        (None, 28, 28, 512)       0         \n",
      "_________________________________________________________________\n",
      "conv_1_31 (Conv2D)           (None, 28, 28, 128)       65664     \n",
      "_________________________________________________________________\n",
      "batch_1_31 (BatchNormalizati (None, 28, 28, 128)       512       \n",
      "_________________________________________________________________\n",
      "act_1_31 (Activation)        (None, 28, 28, 128)       0         \n",
      "_________________________________________________________________\n",
      "conv_1_32 (Conv2D)           (None, 28, 28, 128)       147584    \n",
      "_________________________________________________________________\n",
      "batch_1_32 (BatchNormalizati (None, 28, 28, 128)       512       \n",
      "_________________________________________________________________\n",
      "act_1_32 (Activation)        (None, 28, 28, 128)       0         \n",
      "_________________________________________________________________\n",
      "conv_1_33 (Conv2D)           (None, 28, 28, 512)       66048     \n",
      "_________________________________________________________________\n",
      "batch_1_33 (BatchNormalizati (None, 28, 28, 512)       2048      \n",
      "_________________________________________________________________\n",
      "act_1_33 (Activation)        (None, 28, 28, 512)       0         \n",
      "_________________________________________________________________\n",
      "conv_2_01 (Conv2D)           (None, 14, 14, 256)       131328    \n",
      "_________________________________________________________________\n",
      "batch_2_01 (BatchNormalizati (None, 14, 14, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_2_01 (Activation)        (None, 14, 14, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_2_02 (Conv2D)           (None, 14, 14, 256)       590080    \n",
      "_________________________________________________________________\n",
      "batch_2_02 (BatchNormalizati (None, 14, 14, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_2_02 (Activation)        (None, 14, 14, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_2_03 (Conv2D)           (None, 14, 14, 1024)      263168    \n",
      "_________________________________________________________________\n",
      "batch_2_03 (BatchNormalizati (None, 14, 14, 1024)      4096      \n",
      "_________________________________________________________________\n",
      "act_2_03 (Activation)        (None, 14, 14, 1024)      0         \n",
      "_________________________________________________________________\n",
      "conv_2_11 (Conv2D)           (None, 14, 14, 256)       262400    \n",
      "_________________________________________________________________\n",
      "batch_2_11 (BatchNormalizati (None, 14, 14, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_2_11 (Activation)        (None, 14, 14, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_2_12 (Conv2D)           (None, 14, 14, 256)       590080    \n",
      "_________________________________________________________________\n",
      "batch_2_12 (BatchNormalizati (None, 14, 14, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_2_12 (Activation)        (None, 14, 14, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_2_13 (Conv2D)           (None, 14, 14, 1024)      263168    \n",
      "_________________________________________________________________\n",
      "batch_2_13 (BatchNormalizati (None, 14, 14, 1024)      4096      \n",
      "_________________________________________________________________\n",
      "act_2_13 (Activation)        (None, 14, 14, 1024)      0         \n",
      "_________________________________________________________________\n",
      "conv_2_21 (Conv2D)           (None, 14, 14, 256)       262400    \n",
      "_________________________________________________________________\n",
      "batch_2_21 (BatchNormalizati (None, 14, 14, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_2_21 (Activation)        (None, 14, 14, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_2_22 (Conv2D)           (None, 14, 14, 256)       590080    \n",
      "_________________________________________________________________\n",
      "batch_2_22 (BatchNormalizati (None, 14, 14, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_2_22 (Activation)        (None, 14, 14, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_2_23 (Conv2D)           (None, 14, 14, 1024)      263168    \n",
      "_________________________________________________________________\n",
      "batch_2_23 (BatchNormalizati (None, 14, 14, 1024)      4096      \n",
      "_________________________________________________________________\n",
      "act_2_23 (Activation)        (None, 14, 14, 1024)      0         \n",
      "_________________________________________________________________\n",
      "conv_2_31 (Conv2D)           (None, 14, 14, 256)       262400    \n",
      "_________________________________________________________________\n",
      "batch_2_31 (BatchNormalizati (None, 14, 14, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_2_31 (Activation)        (None, 14, 14, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_2_32 (Conv2D)           (None, 14, 14, 256)       590080    \n",
      "_________________________________________________________________\n",
      "batch_2_32 (BatchNormalizati (None, 14, 14, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_2_32 (Activation)        (None, 14, 14, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_2_33 (Conv2D)           (None, 14, 14, 1024)      263168    \n",
      "_________________________________________________________________\n",
      "batch_2_33 (BatchNormalizati (None, 14, 14, 1024)      4096      \n",
      "_________________________________________________________________\n",
      "act_2_33 (Activation)        (None, 14, 14, 1024)      0         \n",
      "_________________________________________________________________\n",
      "conv_2_41 (Conv2D)           (None, 14, 14, 256)       262400    \n",
      "_________________________________________________________________\n",
      "batch_2_41 (BatchNormalizati (None, 14, 14, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_2_41 (Activation)        (None, 14, 14, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_2_42 (Conv2D)           (None, 14, 14, 256)       590080    \n",
      "_________________________________________________________________\n",
      "batch_2_42 (BatchNormalizati (None, 14, 14, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_2_42 (Activation)        (None, 14, 14, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_2_43 (Conv2D)           (None, 14, 14, 1024)      263168    \n",
      "_________________________________________________________________\n",
      "batch_2_43 (BatchNormalizati (None, 14, 14, 1024)      4096      \n",
      "_________________________________________________________________\n",
      "act_2_43 (Activation)        (None, 14, 14, 1024)      0         \n",
      "_________________________________________________________________\n",
      "conv_2_51 (Conv2D)           (None, 14, 14, 256)       262400    \n",
      "_________________________________________________________________\n",
      "batch_2_51 (BatchNormalizati (None, 14, 14, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_2_51 (Activation)        (None, 14, 14, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_2_52 (Conv2D)           (None, 14, 14, 256)       590080    \n",
      "_________________________________________________________________\n",
      "batch_2_52 (BatchNormalizati (None, 14, 14, 256)       1024      \n",
      "_________________________________________________________________\n",
      "act_2_52 (Activation)        (None, 14, 14, 256)       0         \n",
      "_________________________________________________________________\n",
      "conv_2_53 (Conv2D)           (None, 14, 14, 1024)      263168    \n",
      "_________________________________________________________________\n",
      "batch_2_53 (BatchNormalizati (None, 14, 14, 1024)      4096      \n",
      "_________________________________________________________________\n",
      "act_2_53 (Activation)        (None, 14, 14, 1024)      0         \n",
      "_________________________________________________________________\n",
      "conv_3_01 (Conv2D)           (None, 7, 7, 512)         524800    \n",
      "_________________________________________________________________\n",
      "batch_3_01 (BatchNormalizati (None, 7, 7, 512)         2048      \n",
      "_________________________________________________________________\n",
      "act_3_01 (Activation)        (None, 7, 7, 512)         0         \n",
      "_________________________________________________________________\n",
      "conv_3_02 (Conv2D)           (None, 7, 7, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "batch_3_02 (BatchNormalizati (None, 7, 7, 512)         2048      \n",
      "_________________________________________________________________\n",
      "act_3_02 (Activation)        (None, 7, 7, 512)         0         \n",
      "_________________________________________________________________\n",
      "conv_3_03 (Conv2D)           (None, 7, 7, 2048)        1050624   \n",
      "_________________________________________________________________\n",
      "batch_3_03 (BatchNormalizati (None, 7, 7, 2048)        8192      \n",
      "_________________________________________________________________\n",
      "act_3_03 (Activation)        (None, 7, 7, 2048)        0         \n",
      "_________________________________________________________________\n",
      "conv_3_11 (Conv2D)           (None, 7, 7, 512)         1049088   \n",
      "_________________________________________________________________\n",
      "batch_3_11 (BatchNormalizati (None, 7, 7, 512)         2048      \n",
      "_________________________________________________________________\n",
      "act_3_11 (Activation)        (None, 7, 7, 512)         0         \n",
      "_________________________________________________________________\n",
      "conv_3_12 (Conv2D)           (None, 7, 7, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "batch_3_12 (BatchNormalizati (None, 7, 7, 512)         2048      \n",
      "_________________________________________________________________\n",
      "act_3_12 (Activation)        (None, 7, 7, 512)         0         \n",
      "_________________________________________________________________\n",
      "conv_3_13 (Conv2D)           (None, 7, 7, 2048)        1050624   \n",
      "_________________________________________________________________\n",
      "batch_3_13 (BatchNormalizati (None, 7, 7, 2048)        8192      \n",
      "_________________________________________________________________\n",
      "act_3_13 (Activation)        (None, 7, 7, 2048)        0         \n",
      "_________________________________________________________________\n",
      "conv_3_21 (Conv2D)           (None, 7, 7, 512)         1049088   \n",
      "_________________________________________________________________\n",
      "batch_3_21 (BatchNormalizati (None, 7, 7, 512)         2048      \n",
      "_________________________________________________________________\n",
      "act_3_21 (Activation)        (None, 7, 7, 512)         0         \n",
      "_________________________________________________________________\n",
      "conv_3_22 (Conv2D)           (None, 7, 7, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "batch_3_22 (BatchNormalizati (None, 7, 7, 512)         2048      \n",
      "_________________________________________________________________\n",
      "act_3_22 (Activation)        (None, 7, 7, 512)         0         \n",
      "_________________________________________________________________\n",
      "conv_3_23 (Conv2D)           (None, 7, 7, 2048)        1050624   \n",
      "_________________________________________________________________\n",
      "batch_3_23 (BatchNormalizati (None, 7, 7, 2048)        8192      \n",
      "_________________________________________________________________\n",
      "act_3_23 (Activation)        (None, 7, 7, 2048)        0         \n",
      "_________________________________________________________________\n",
      "GAP (GlobalAveragePooling2D) (None, 2048)              0         \n",
      "_________________________________________________________________\n",
      "FC (Dense)                   (None, 10)                20490     \n",
      "=================================================================\n",
      "Total params: 20,820,106\n",
      "Trainable params: 20,774,666\n",
      "Non-trainable params: 45,440\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "plain_50 = build_plain(num_cnn_list = [3, 4, 6, 3],\n",
    "                       channel_list=[64, 128, 256, 512],\n",
    "                       is_50=True,\n",
    "                       input_shape=(224,224,3))\n",
    "plain_50.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ResNet"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Residual block"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def build_residual_block(input_layer, block_num=0, num_cnn=3, channel=64, is_50=True):\n",
    "    x = input_layer\n",
    "    \n",
    "    # ResNet 50\n",
    "    if is_50:\n",
    "        for i in range(num_cnn):\n",
    "            if i==0:\n",
    "                shortcut = conv_batch(x, channel*4, (1,1), strides=2, activation=None, name=f'_shortcut_{block_num}_{i}')\n",
    "                x = conv_batch(x, channel, (1,1), strides=2, name=f'{block_num}_{i}1')\n",
    "                x = conv_batch(x, channel, (3,3), name=f'{block_num}_{i}2')\n",
    "                x = conv_batch(x, channel*4, (1,1), activation=None, name=f'{block_num}_{i}3')\n",
    "            else:\n",
    "                shortcut = x\n",
    "                x = conv_batch(x, channel, (1,1), name=f'{block_num}_{i}1')\n",
    "                x = conv_batch(x, channel, (3,3), name=f'{block_num}_{i}2')\n",
    "                x = conv_batch(x, channel*4, (1,1), activation=None, name=f'{block_num}_{i}3')\n",
    "            \n",
    "            x = keras.layers.Add(name=f'shortcut_add_{block_num}_{i}')([x, shortcut])\n",
    "            x = keras.layers.Activation('relu', name=f'shortcut_act_{block_num}_{i}')(x)\n",
    "    # ResNet 34\n",
    "    else:\n",
    "        for i in range(num_cnn):\n",
    "            if block_num > 0 and i==0:\n",
    "                shortcut = conv_batch(x, channel, (1,1), strides=2, activation=None, name=f'_shortcut_{block_num}_{i}')\n",
    "                x = conv_batch(x, channel, (3,3), strides=2, name=f'{block_num}_{i}1')\n",
    "                x = conv_batch(x, channel, (3,3), activation=None, name=f'{block_num}_{i}2')\n",
    "            else:\n",
    "                shortcut = x\n",
    "                x = conv_batch(x, channel, (3,3), name=f'{block_num}_{i}1')\n",
    "                x = conv_batch(x, channel, (3,3), activation=None, name=f'{block_num}_{i}2')\n",
    "            \n",
    "            x = keras.layers.Add(name=f'shortcut_add_{block_num}_{i}')([x, shortcut])\n",
    "            x = keras.layers.Activation('relu', name=f'shortcut_act_{block_num}_{i}')(x)\n",
    "    \n",
    "\n",
    "    return x"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Build ResNet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def build_resnet(num_cnn_list, channel_list, is_50=True, cats_vs_dogs=False, input_shape=(32,32,3), num_classes=10, name='ResNet_50'):\n",
    "    \n",
    "    assert len(num_cnn_list) == len(channel_list)\n",
    "    \n",
    "    input_layer = keras.layers.Input(shape=input_shape, name='Input')\n",
    "    \n",
    "    # first layer\n",
    "    conv1 = conv_batch(input_layer, 64, (7,7), strides=2, name='first')\n",
    "    res = keras.layers.MaxPool2D(pool_size=(2,2), strides=2, name='maxpool_first')(conv1)\n",
    "    \n",
    "    # Residual block\n",
    "    for block_num, (cnn, channel) in enumerate(zip(num_cnn_list, channel_list)):\n",
    "        res = build_residual_block(res, block_num, cnn, channel, is_50)\n",
    "        \n",
    "    gap = keras.layers.GlobalAveragePooling2D(name='GAP')(res)\n",
    "    if cats_vs_dogs:\n",
    "        output = keras.layers.Dense(1, activation='sigmoid',kernel_initializer='he_normal', name='FC')(gap)\n",
    "    else:\n",
    "        output = keras.layers.Dense(num_classes, activation='softmax',kernel_initializer='he_normal', name='FC')(gap)\n",
    "    \n",
    "    model = keras.Model(inputs=input_layer, outputs=output, name=name)\n",
    "    return model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ResNet-34"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"ResNet_34\"\n",
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "Input (InputLayer)              [(None, 224, 224, 3) 0                                            \n",
      "__________________________________________________________________________________________________\n",
      "conv_first (Conv2D)             (None, 112, 112, 64) 9472        Input[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "batch_first (BatchNormalization (None, 112, 112, 64) 256         conv_first[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "act_first (Activation)          (None, 112, 112, 64) 0           batch_first[0][0]                \n",
      "__________________________________________________________________________________________________\n",
      "maxpool_first (MaxPooling2D)    (None, 56, 56, 64)   0           act_first[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "conv_0_01 (Conv2D)              (None, 56, 56, 64)   36928       maxpool_first[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_0_01 (BatchNormalization) (None, 56, 56, 64)   256         conv_0_01[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_0_01 (Activation)           (None, 56, 56, 64)   0           batch_0_01[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_0_02 (Conv2D)              (None, 56, 56, 64)   36928       act_0_01[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_0_02 (BatchNormalization) (None, 56, 56, 64)   256         conv_0_02[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_0_0 (Add)          (None, 56, 56, 64)   0           batch_0_02[0][0]                 \n",
      "                                                                 maxpool_first[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_0_0 (Activation)   (None, 56, 56, 64)   0           shortcut_add_0_0[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_0_11 (Conv2D)              (None, 56, 56, 64)   36928       shortcut_act_0_0[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_0_11 (BatchNormalization) (None, 56, 56, 64)   256         conv_0_11[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_0_11 (Activation)           (None, 56, 56, 64)   0           batch_0_11[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_0_12 (Conv2D)              (None, 56, 56, 64)   36928       act_0_11[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_0_12 (BatchNormalization) (None, 56, 56, 64)   256         conv_0_12[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_0_1 (Add)          (None, 56, 56, 64)   0           batch_0_12[0][0]                 \n",
      "                                                                 shortcut_act_0_0[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_0_1 (Activation)   (None, 56, 56, 64)   0           shortcut_add_0_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_0_21 (Conv2D)              (None, 56, 56, 64)   36928       shortcut_act_0_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_0_21 (BatchNormalization) (None, 56, 56, 64)   256         conv_0_21[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_0_21 (Activation)           (None, 56, 56, 64)   0           batch_0_21[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_0_22 (Conv2D)              (None, 56, 56, 64)   36928       act_0_21[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_0_22 (BatchNormalization) (None, 56, 56, 64)   256         conv_0_22[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_0_2 (Add)          (None, 56, 56, 64)   0           batch_0_22[0][0]                 \n",
      "                                                                 shortcut_act_0_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_0_2 (Activation)   (None, 56, 56, 64)   0           shortcut_add_0_2[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_1_01 (Conv2D)              (None, 28, 28, 128)  73856       shortcut_act_0_2[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_1_01 (BatchNormalization) (None, 28, 28, 128)  512         conv_1_01[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_1_01 (Activation)           (None, 28, 28, 128)  0           batch_1_01[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_1_02 (Conv2D)              (None, 28, 28, 128)  147584      act_1_01[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "conv__shortcut_1_0 (Conv2D)     (None, 28, 28, 128)  8320        shortcut_act_0_2[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_1_02 (BatchNormalization) (None, 28, 28, 128)  512         conv_1_02[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch__shortcut_1_0 (BatchNorma (None, 28, 28, 128)  512         conv__shortcut_1_0[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_1_0 (Add)          (None, 28, 28, 128)  0           batch_1_02[0][0]                 \n",
      "                                                                 batch__shortcut_1_0[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_1_0 (Activation)   (None, 28, 28, 128)  0           shortcut_add_1_0[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_1_11 (Conv2D)              (None, 28, 28, 128)  147584      shortcut_act_1_0[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_1_11 (BatchNormalization) (None, 28, 28, 128)  512         conv_1_11[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_1_11 (Activation)           (None, 28, 28, 128)  0           batch_1_11[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_1_12 (Conv2D)              (None, 28, 28, 128)  147584      act_1_11[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_1_12 (BatchNormalization) (None, 28, 28, 128)  512         conv_1_12[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_1_1 (Add)          (None, 28, 28, 128)  0           batch_1_12[0][0]                 \n",
      "                                                                 shortcut_act_1_0[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_1_1 (Activation)   (None, 28, 28, 128)  0           shortcut_add_1_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_1_21 (Conv2D)              (None, 28, 28, 128)  147584      shortcut_act_1_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_1_21 (BatchNormalization) (None, 28, 28, 128)  512         conv_1_21[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_1_21 (Activation)           (None, 28, 28, 128)  0           batch_1_21[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_1_22 (Conv2D)              (None, 28, 28, 128)  147584      act_1_21[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_1_22 (BatchNormalization) (None, 28, 28, 128)  512         conv_1_22[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_1_2 (Add)          (None, 28, 28, 128)  0           batch_1_22[0][0]                 \n",
      "                                                                 shortcut_act_1_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_1_2 (Activation)   (None, 28, 28, 128)  0           shortcut_add_1_2[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_1_31 (Conv2D)              (None, 28, 28, 128)  147584      shortcut_act_1_2[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_1_31 (BatchNormalization) (None, 28, 28, 128)  512         conv_1_31[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_1_31 (Activation)           (None, 28, 28, 128)  0           batch_1_31[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_1_32 (Conv2D)              (None, 28, 28, 128)  147584      act_1_31[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_1_32 (BatchNormalization) (None, 28, 28, 128)  512         conv_1_32[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_1_3 (Add)          (None, 28, 28, 128)  0           batch_1_32[0][0]                 \n",
      "                                                                 shortcut_act_1_2[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_1_3 (Activation)   (None, 28, 28, 128)  0           shortcut_add_1_3[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_01 (Conv2D)              (None, 14, 14, 256)  295168      shortcut_act_1_3[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_01 (BatchNormalization) (None, 14, 14, 256)  1024        conv_2_01[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_2_01 (Activation)           (None, 14, 14, 256)  0           batch_2_01[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_02 (Conv2D)              (None, 14, 14, 256)  590080      act_2_01[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "conv__shortcut_2_0 (Conv2D)     (None, 14, 14, 256)  33024       shortcut_act_1_3[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_02 (BatchNormalization) (None, 14, 14, 256)  1024        conv_2_02[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch__shortcut_2_0 (BatchNorma (None, 14, 14, 256)  1024        conv__shortcut_2_0[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_2_0 (Add)          (None, 14, 14, 256)  0           batch_2_02[0][0]                 \n",
      "                                                                 batch__shortcut_2_0[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_2_0 (Activation)   (None, 14, 14, 256)  0           shortcut_add_2_0[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_11 (Conv2D)              (None, 14, 14, 256)  590080      shortcut_act_2_0[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_11 (BatchNormalization) (None, 14, 14, 256)  1024        conv_2_11[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_2_11 (Activation)           (None, 14, 14, 256)  0           batch_2_11[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_12 (Conv2D)              (None, 14, 14, 256)  590080      act_2_11[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_12 (BatchNormalization) (None, 14, 14, 256)  1024        conv_2_12[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_2_1 (Add)          (None, 14, 14, 256)  0           batch_2_12[0][0]                 \n",
      "                                                                 shortcut_act_2_0[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_2_1 (Activation)   (None, 14, 14, 256)  0           shortcut_add_2_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_21 (Conv2D)              (None, 14, 14, 256)  590080      shortcut_act_2_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_21 (BatchNormalization) (None, 14, 14, 256)  1024        conv_2_21[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_2_21 (Activation)           (None, 14, 14, 256)  0           batch_2_21[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_22 (Conv2D)              (None, 14, 14, 256)  590080      act_2_21[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_22 (BatchNormalization) (None, 14, 14, 256)  1024        conv_2_22[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_2_2 (Add)          (None, 14, 14, 256)  0           batch_2_22[0][0]                 \n",
      "                                                                 shortcut_act_2_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_2_2 (Activation)   (None, 14, 14, 256)  0           shortcut_add_2_2[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_31 (Conv2D)              (None, 14, 14, 256)  590080      shortcut_act_2_2[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_31 (BatchNormalization) (None, 14, 14, 256)  1024        conv_2_31[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_2_31 (Activation)           (None, 14, 14, 256)  0           batch_2_31[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_32 (Conv2D)              (None, 14, 14, 256)  590080      act_2_31[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_32 (BatchNormalization) (None, 14, 14, 256)  1024        conv_2_32[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_2_3 (Add)          (None, 14, 14, 256)  0           batch_2_32[0][0]                 \n",
      "                                                                 shortcut_act_2_2[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_2_3 (Activation)   (None, 14, 14, 256)  0           shortcut_add_2_3[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_41 (Conv2D)              (None, 14, 14, 256)  590080      shortcut_act_2_3[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_41 (BatchNormalization) (None, 14, 14, 256)  1024        conv_2_41[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_2_41 (Activation)           (None, 14, 14, 256)  0           batch_2_41[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_42 (Conv2D)              (None, 14, 14, 256)  590080      act_2_41[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_42 (BatchNormalization) (None, 14, 14, 256)  1024        conv_2_42[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_2_4 (Add)          (None, 14, 14, 256)  0           batch_2_42[0][0]                 \n",
      "                                                                 shortcut_act_2_3[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_2_4 (Activation)   (None, 14, 14, 256)  0           shortcut_add_2_4[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_51 (Conv2D)              (None, 14, 14, 256)  590080      shortcut_act_2_4[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_51 (BatchNormalization) (None, 14, 14, 256)  1024        conv_2_51[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_2_51 (Activation)           (None, 14, 14, 256)  0           batch_2_51[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_52 (Conv2D)              (None, 14, 14, 256)  590080      act_2_51[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_52 (BatchNormalization) (None, 14, 14, 256)  1024        conv_2_52[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_2_5 (Add)          (None, 14, 14, 256)  0           batch_2_52[0][0]                 \n",
      "                                                                 shortcut_act_2_4[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_2_5 (Activation)   (None, 14, 14, 256)  0           shortcut_add_2_5[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_3_01 (Conv2D)              (None, 7, 7, 512)    1180160     shortcut_act_2_5[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_3_01 (BatchNormalization) (None, 7, 7, 512)    2048        conv_3_01[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_3_01 (Activation)           (None, 7, 7, 512)    0           batch_3_01[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_3_02 (Conv2D)              (None, 7, 7, 512)    2359808     act_3_01[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "conv__shortcut_3_0 (Conv2D)     (None, 7, 7, 512)    131584      shortcut_act_2_5[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_3_02 (BatchNormalization) (None, 7, 7, 512)    2048        conv_3_02[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch__shortcut_3_0 (BatchNorma (None, 7, 7, 512)    2048        conv__shortcut_3_0[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_3_0 (Add)          (None, 7, 7, 512)    0           batch_3_02[0][0]                 \n",
      "                                                                 batch__shortcut_3_0[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_3_0 (Activation)   (None, 7, 7, 512)    0           shortcut_add_3_0[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_3_11 (Conv2D)              (None, 7, 7, 512)    2359808     shortcut_act_3_0[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_3_11 (BatchNormalization) (None, 7, 7, 512)    2048        conv_3_11[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_3_11 (Activation)           (None, 7, 7, 512)    0           batch_3_11[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_3_12 (Conv2D)              (None, 7, 7, 512)    2359808     act_3_11[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_3_12 (BatchNormalization) (None, 7, 7, 512)    2048        conv_3_12[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_3_1 (Add)          (None, 7, 7, 512)    0           batch_3_12[0][0]                 \n",
      "                                                                 shortcut_act_3_0[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_3_1 (Activation)   (None, 7, 7, 512)    0           shortcut_add_3_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_3_21 (Conv2D)              (None, 7, 7, 512)    2359808     shortcut_act_3_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_3_21 (BatchNormalization) (None, 7, 7, 512)    2048        conv_3_21[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_3_21 (Activation)           (None, 7, 7, 512)    0           batch_3_21[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_3_22 (Conv2D)              (None, 7, 7, 512)    2359808     act_3_21[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_3_22 (BatchNormalization) (None, 7, 7, 512)    2048        conv_3_22[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_3_2 (Add)          (None, 7, 7, 512)    0           batch_3_22[0][0]                 \n",
      "                                                                 shortcut_act_3_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_3_2 (Activation)   (None, 7, 7, 512)    0           shortcut_add_3_2[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "GAP (GlobalAveragePooling2D)    (None, 512)          0           shortcut_act_3_2[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "FC (Dense)                      (None, 10)           5130        GAP[0][0]                        \n",
      "==================================================================================================\n",
      "Total params: 21,315,338\n",
      "Trainable params: 21,298,314\n",
      "Non-trainable params: 17,024\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "resnet_34 = build_resnet(num_cnn_list = [3, 4, 6, 3],\n",
    "                         channel_list=[64, 128, 256, 512],\n",
    "                         is_50=False,\n",
    "                         input_shape=(224,224,3),\n",
    "                         name='ResNet_34')\n",
    "resnet_34.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ResNet-50"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"ResNet_50\"\n",
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "Input (InputLayer)              [(None, 224, 224, 3) 0                                            \n",
      "__________________________________________________________________________________________________\n",
      "conv_first (Conv2D)             (None, 112, 112, 64) 9472        Input[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "batch_first (BatchNormalization (None, 112, 112, 64) 256         conv_first[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "act_first (Activation)          (None, 112, 112, 64) 0           batch_first[0][0]                \n",
      "__________________________________________________________________________________________________\n",
      "maxpool_first (MaxPooling2D)    (None, 56, 56, 64)   0           act_first[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "conv_0_01 (Conv2D)              (None, 28, 28, 64)   4160        maxpool_first[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_0_01 (BatchNormalization) (None, 28, 28, 64)   256         conv_0_01[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_0_01 (Activation)           (None, 28, 28, 64)   0           batch_0_01[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_0_02 (Conv2D)              (None, 28, 28, 64)   36928       act_0_01[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_0_02 (BatchNormalization) (None, 28, 28, 64)   256         conv_0_02[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_0_02 (Activation)           (None, 28, 28, 64)   0           batch_0_02[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_0_03 (Conv2D)              (None, 28, 28, 256)  16640       act_0_02[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "conv__shortcut_0_0 (Conv2D)     (None, 28, 28, 256)  16640       maxpool_first[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_0_03 (BatchNormalization) (None, 28, 28, 256)  1024        conv_0_03[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch__shortcut_0_0 (BatchNorma (None, 28, 28, 256)  1024        conv__shortcut_0_0[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_0_0 (Add)          (None, 28, 28, 256)  0           batch_0_03[0][0]                 \n",
      "                                                                 batch__shortcut_0_0[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_0_0 (Activation)   (None, 28, 28, 256)  0           shortcut_add_0_0[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_0_11 (Conv2D)              (None, 28, 28, 64)   16448       shortcut_act_0_0[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_0_11 (BatchNormalization) (None, 28, 28, 64)   256         conv_0_11[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_0_11 (Activation)           (None, 28, 28, 64)   0           batch_0_11[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_0_12 (Conv2D)              (None, 28, 28, 64)   36928       act_0_11[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_0_12 (BatchNormalization) (None, 28, 28, 64)   256         conv_0_12[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_0_12 (Activation)           (None, 28, 28, 64)   0           batch_0_12[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_0_13 (Conv2D)              (None, 28, 28, 256)  16640       act_0_12[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_0_13 (BatchNormalization) (None, 28, 28, 256)  1024        conv_0_13[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_0_1 (Add)          (None, 28, 28, 256)  0           batch_0_13[0][0]                 \n",
      "                                                                 shortcut_act_0_0[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_0_1 (Activation)   (None, 28, 28, 256)  0           shortcut_add_0_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_0_21 (Conv2D)              (None, 28, 28, 64)   16448       shortcut_act_0_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_0_21 (BatchNormalization) (None, 28, 28, 64)   256         conv_0_21[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_0_21 (Activation)           (None, 28, 28, 64)   0           batch_0_21[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_0_22 (Conv2D)              (None, 28, 28, 64)   36928       act_0_21[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_0_22 (BatchNormalization) (None, 28, 28, 64)   256         conv_0_22[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_0_22 (Activation)           (None, 28, 28, 64)   0           batch_0_22[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_0_23 (Conv2D)              (None, 28, 28, 256)  16640       act_0_22[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_0_23 (BatchNormalization) (None, 28, 28, 256)  1024        conv_0_23[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_0_2 (Add)          (None, 28, 28, 256)  0           batch_0_23[0][0]                 \n",
      "                                                                 shortcut_act_0_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_0_2 (Activation)   (None, 28, 28, 256)  0           shortcut_add_0_2[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_1_01 (Conv2D)              (None, 14, 14, 128)  32896       shortcut_act_0_2[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_1_01 (BatchNormalization) (None, 14, 14, 128)  512         conv_1_01[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_1_01 (Activation)           (None, 14, 14, 128)  0           batch_1_01[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_1_02 (Conv2D)              (None, 14, 14, 128)  147584      act_1_01[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_1_02 (BatchNormalization) (None, 14, 14, 128)  512         conv_1_02[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_1_02 (Activation)           (None, 14, 14, 128)  0           batch_1_02[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_1_03 (Conv2D)              (None, 14, 14, 512)  66048       act_1_02[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "conv__shortcut_1_0 (Conv2D)     (None, 14, 14, 512)  131584      shortcut_act_0_2[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_1_03 (BatchNormalization) (None, 14, 14, 512)  2048        conv_1_03[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch__shortcut_1_0 (BatchNorma (None, 14, 14, 512)  2048        conv__shortcut_1_0[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_1_0 (Add)          (None, 14, 14, 512)  0           batch_1_03[0][0]                 \n",
      "                                                                 batch__shortcut_1_0[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_1_0 (Activation)   (None, 14, 14, 512)  0           shortcut_add_1_0[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_1_11 (Conv2D)              (None, 14, 14, 128)  65664       shortcut_act_1_0[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_1_11 (BatchNormalization) (None, 14, 14, 128)  512         conv_1_11[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_1_11 (Activation)           (None, 14, 14, 128)  0           batch_1_11[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_1_12 (Conv2D)              (None, 14, 14, 128)  147584      act_1_11[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_1_12 (BatchNormalization) (None, 14, 14, 128)  512         conv_1_12[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_1_12 (Activation)           (None, 14, 14, 128)  0           batch_1_12[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_1_13 (Conv2D)              (None, 14, 14, 512)  66048       act_1_12[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_1_13 (BatchNormalization) (None, 14, 14, 512)  2048        conv_1_13[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_1_1 (Add)          (None, 14, 14, 512)  0           batch_1_13[0][0]                 \n",
      "                                                                 shortcut_act_1_0[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_1_1 (Activation)   (None, 14, 14, 512)  0           shortcut_add_1_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_1_21 (Conv2D)              (None, 14, 14, 128)  65664       shortcut_act_1_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_1_21 (BatchNormalization) (None, 14, 14, 128)  512         conv_1_21[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_1_21 (Activation)           (None, 14, 14, 128)  0           batch_1_21[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_1_22 (Conv2D)              (None, 14, 14, 128)  147584      act_1_21[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_1_22 (BatchNormalization) (None, 14, 14, 128)  512         conv_1_22[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_1_22 (Activation)           (None, 14, 14, 128)  0           batch_1_22[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_1_23 (Conv2D)              (None, 14, 14, 512)  66048       act_1_22[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_1_23 (BatchNormalization) (None, 14, 14, 512)  2048        conv_1_23[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_1_2 (Add)          (None, 14, 14, 512)  0           batch_1_23[0][0]                 \n",
      "                                                                 shortcut_act_1_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_1_2 (Activation)   (None, 14, 14, 512)  0           shortcut_add_1_2[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_1_31 (Conv2D)              (None, 14, 14, 128)  65664       shortcut_act_1_2[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_1_31 (BatchNormalization) (None, 14, 14, 128)  512         conv_1_31[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_1_31 (Activation)           (None, 14, 14, 128)  0           batch_1_31[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_1_32 (Conv2D)              (None, 14, 14, 128)  147584      act_1_31[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_1_32 (BatchNormalization) (None, 14, 14, 128)  512         conv_1_32[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_1_32 (Activation)           (None, 14, 14, 128)  0           batch_1_32[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_1_33 (Conv2D)              (None, 14, 14, 512)  66048       act_1_32[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_1_33 (BatchNormalization) (None, 14, 14, 512)  2048        conv_1_33[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_1_3 (Add)          (None, 14, 14, 512)  0           batch_1_33[0][0]                 \n",
      "                                                                 shortcut_act_1_2[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_1_3 (Activation)   (None, 14, 14, 512)  0           shortcut_add_1_3[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_01 (Conv2D)              (None, 7, 7, 256)    131328      shortcut_act_1_3[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_01 (BatchNormalization) (None, 7, 7, 256)    1024        conv_2_01[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_2_01 (Activation)           (None, 7, 7, 256)    0           batch_2_01[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_02 (Conv2D)              (None, 7, 7, 256)    590080      act_2_01[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_02 (BatchNormalization) (None, 7, 7, 256)    1024        conv_2_02[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_2_02 (Activation)           (None, 7, 7, 256)    0           batch_2_02[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_03 (Conv2D)              (None, 7, 7, 1024)   263168      act_2_02[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "conv__shortcut_2_0 (Conv2D)     (None, 7, 7, 1024)   525312      shortcut_act_1_3[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_03 (BatchNormalization) (None, 7, 7, 1024)   4096        conv_2_03[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch__shortcut_2_0 (BatchNorma (None, 7, 7, 1024)   4096        conv__shortcut_2_0[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_2_0 (Add)          (None, 7, 7, 1024)   0           batch_2_03[0][0]                 \n",
      "                                                                 batch__shortcut_2_0[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_2_0 (Activation)   (None, 7, 7, 1024)   0           shortcut_add_2_0[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_11 (Conv2D)              (None, 7, 7, 256)    262400      shortcut_act_2_0[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_11 (BatchNormalization) (None, 7, 7, 256)    1024        conv_2_11[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_2_11 (Activation)           (None, 7, 7, 256)    0           batch_2_11[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_12 (Conv2D)              (None, 7, 7, 256)    590080      act_2_11[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_12 (BatchNormalization) (None, 7, 7, 256)    1024        conv_2_12[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_2_12 (Activation)           (None, 7, 7, 256)    0           batch_2_12[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_13 (Conv2D)              (None, 7, 7, 1024)   263168      act_2_12[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_13 (BatchNormalization) (None, 7, 7, 1024)   4096        conv_2_13[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_2_1 (Add)          (None, 7, 7, 1024)   0           batch_2_13[0][0]                 \n",
      "                                                                 shortcut_act_2_0[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_2_1 (Activation)   (None, 7, 7, 1024)   0           shortcut_add_2_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_21 (Conv2D)              (None, 7, 7, 256)    262400      shortcut_act_2_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_21 (BatchNormalization) (None, 7, 7, 256)    1024        conv_2_21[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_2_21 (Activation)           (None, 7, 7, 256)    0           batch_2_21[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_22 (Conv2D)              (None, 7, 7, 256)    590080      act_2_21[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_22 (BatchNormalization) (None, 7, 7, 256)    1024        conv_2_22[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_2_22 (Activation)           (None, 7, 7, 256)    0           batch_2_22[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_23 (Conv2D)              (None, 7, 7, 1024)   263168      act_2_22[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_23 (BatchNormalization) (None, 7, 7, 1024)   4096        conv_2_23[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_2_2 (Add)          (None, 7, 7, 1024)   0           batch_2_23[0][0]                 \n",
      "                                                                 shortcut_act_2_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_2_2 (Activation)   (None, 7, 7, 1024)   0           shortcut_add_2_2[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_31 (Conv2D)              (None, 7, 7, 256)    262400      shortcut_act_2_2[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_31 (BatchNormalization) (None, 7, 7, 256)    1024        conv_2_31[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_2_31 (Activation)           (None, 7, 7, 256)    0           batch_2_31[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_32 (Conv2D)              (None, 7, 7, 256)    590080      act_2_31[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_32 (BatchNormalization) (None, 7, 7, 256)    1024        conv_2_32[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_2_32 (Activation)           (None, 7, 7, 256)    0           batch_2_32[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_33 (Conv2D)              (None, 7, 7, 1024)   263168      act_2_32[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_33 (BatchNormalization) (None, 7, 7, 1024)   4096        conv_2_33[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_2_3 (Add)          (None, 7, 7, 1024)   0           batch_2_33[0][0]                 \n",
      "                                                                 shortcut_act_2_2[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_2_3 (Activation)   (None, 7, 7, 1024)   0           shortcut_add_2_3[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_41 (Conv2D)              (None, 7, 7, 256)    262400      shortcut_act_2_3[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_41 (BatchNormalization) (None, 7, 7, 256)    1024        conv_2_41[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_2_41 (Activation)           (None, 7, 7, 256)    0           batch_2_41[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_42 (Conv2D)              (None, 7, 7, 256)    590080      act_2_41[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_42 (BatchNormalization) (None, 7, 7, 256)    1024        conv_2_42[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_2_42 (Activation)           (None, 7, 7, 256)    0           batch_2_42[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_43 (Conv2D)              (None, 7, 7, 1024)   263168      act_2_42[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_43 (BatchNormalization) (None, 7, 7, 1024)   4096        conv_2_43[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_2_4 (Add)          (None, 7, 7, 1024)   0           batch_2_43[0][0]                 \n",
      "                                                                 shortcut_act_2_3[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_2_4 (Activation)   (None, 7, 7, 1024)   0           shortcut_add_2_4[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_51 (Conv2D)              (None, 7, 7, 256)    262400      shortcut_act_2_4[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_51 (BatchNormalization) (None, 7, 7, 256)    1024        conv_2_51[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_2_51 (Activation)           (None, 7, 7, 256)    0           batch_2_51[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_52 (Conv2D)              (None, 7, 7, 256)    590080      act_2_51[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_52 (BatchNormalization) (None, 7, 7, 256)    1024        conv_2_52[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_2_52 (Activation)           (None, 7, 7, 256)    0           batch_2_52[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_2_53 (Conv2D)              (None, 7, 7, 1024)   263168      act_2_52[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_2_53 (BatchNormalization) (None, 7, 7, 1024)   4096        conv_2_53[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_2_5 (Add)          (None, 7, 7, 1024)   0           batch_2_53[0][0]                 \n",
      "                                                                 shortcut_act_2_4[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_2_5 (Activation)   (None, 7, 7, 1024)   0           shortcut_add_2_5[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_3_01 (Conv2D)              (None, 4, 4, 512)    524800      shortcut_act_2_5[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_3_01 (BatchNormalization) (None, 4, 4, 512)    2048        conv_3_01[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_3_01 (Activation)           (None, 4, 4, 512)    0           batch_3_01[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_3_02 (Conv2D)              (None, 4, 4, 512)    2359808     act_3_01[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_3_02 (BatchNormalization) (None, 4, 4, 512)    2048        conv_3_02[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_3_02 (Activation)           (None, 4, 4, 512)    0           batch_3_02[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_3_03 (Conv2D)              (None, 4, 4, 2048)   1050624     act_3_02[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "conv__shortcut_3_0 (Conv2D)     (None, 4, 4, 2048)   2099200     shortcut_act_2_5[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_3_03 (BatchNormalization) (None, 4, 4, 2048)   8192        conv_3_03[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch__shortcut_3_0 (BatchNorma (None, 4, 4, 2048)   8192        conv__shortcut_3_0[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_3_0 (Add)          (None, 4, 4, 2048)   0           batch_3_03[0][0]                 \n",
      "                                                                 batch__shortcut_3_0[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_3_0 (Activation)   (None, 4, 4, 2048)   0           shortcut_add_3_0[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_3_11 (Conv2D)              (None, 4, 4, 512)    1049088     shortcut_act_3_0[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_3_11 (BatchNormalization) (None, 4, 4, 512)    2048        conv_3_11[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_3_11 (Activation)           (None, 4, 4, 512)    0           batch_3_11[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_3_12 (Conv2D)              (None, 4, 4, 512)    2359808     act_3_11[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_3_12 (BatchNormalization) (None, 4, 4, 512)    2048        conv_3_12[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_3_12 (Activation)           (None, 4, 4, 512)    0           batch_3_12[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_3_13 (Conv2D)              (None, 4, 4, 2048)   1050624     act_3_12[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_3_13 (BatchNormalization) (None, 4, 4, 2048)   8192        conv_3_13[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_3_1 (Add)          (None, 4, 4, 2048)   0           batch_3_13[0][0]                 \n",
      "                                                                 shortcut_act_3_0[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_3_1 (Activation)   (None, 4, 4, 2048)   0           shortcut_add_3_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv_3_21 (Conv2D)              (None, 4, 4, 512)    1049088     shortcut_act_3_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "batch_3_21 (BatchNormalization) (None, 4, 4, 512)    2048        conv_3_21[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_3_21 (Activation)           (None, 4, 4, 512)    0           batch_3_21[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_3_22 (Conv2D)              (None, 4, 4, 512)    2359808     act_3_21[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_3_22 (BatchNormalization) (None, 4, 4, 512)    2048        conv_3_22[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "act_3_22 (Activation)           (None, 4, 4, 512)    0           batch_3_22[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv_3_23 (Conv2D)              (None, 4, 4, 2048)   1050624     act_3_22[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_3_23 (BatchNormalization) (None, 4, 4, 2048)   8192        conv_3_23[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_add_3_2 (Add)          (None, 4, 4, 2048)   0           batch_3_23[0][0]                 \n",
      "                                                                 shortcut_act_3_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "shortcut_act_3_2 (Activation)   (None, 4, 4, 2048)   0           shortcut_add_3_2[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "GAP (GlobalAveragePooling2D)    (None, 2048)         0           shortcut_act_3_2[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "FC (Dense)                      (None, 10)           20490       GAP[0][0]                        \n",
      "==================================================================================================\n",
      "Total params: 23,608,202\n",
      "Trainable params: 23,555,082\n",
      "Non-trainable params: 53,120\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "resnet_50 = build_resnet(num_cnn_list = [3, 4, 6, 3],\n",
    "                         channel_list=[64, 128, 256, 512],\n",
    "                         is_50=True,\n",
    "                         input_shape=(224,224,3))\n",
    "resnet_50.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ResNet-34 vs ResNet-50 vs Plain-34 vs Plain-50"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Dataset (cats_and_dog)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "(ds_train, ds_test), ds_info = tfds.load(\n",
    "    'cats_vs_dogs',\n",
    "    split=['train[:80%]', 'train[80%:]'],\n",
    "    shuffle_files=True,\n",
    "    with_info=True,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "FeaturesDict({\n",
      "    'image': Image(shape=(None, None, 3), dtype=tf.uint8),\n",
      "    'image/filename': Text(shape=(), dtype=tf.string),\n",
      "    'label': ClassLabel(shape=(), dtype=tf.int64, num_classes=2),\n",
      "})\n"
     ]
    }
   ],
   "source": [
    "print(ds_info.features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tf.Tensor(18610, shape=(), dtype=int64)\n",
      "tf.Tensor(4652, shape=(), dtype=int64)\n"
     ]
    }
   ],
   "source": [
    "print(tf.data.experimental.cardinality(ds_train))\n",
    "print(tf.data.experimental.cardinality(ds_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "def normalize_and_resize_img(image, label):\n",
    "    image = tf.image.resize(image, (224,224)) # resize\n",
    "    image = tf.cast(image, tf.float32) / 255.\n",
    "    return image, label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "def apply_normalize_on_dataset(ds, is_test=False, batch_size=16):\n",
    "    ds = ds.map(\n",
    "        normalize_and_resize_img, \n",
    "        num_parallel_calls=1\n",
    "    )\n",
    "    ds = ds.batch(batch_size)\n",
    "    if not is_test:\n",
    "        ds = ds.repeat()\n",
    "        ds = ds.shuffle(200)\n",
    "    ds = ds.prefetch(tf.data.experimental.AUTOTUNE)\n",
    "    return ds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x648 with 9 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = tfds.show_examples(ds_train, ds_info)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x648 with 9 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = tfds.show_examples(ds_test, ds_info)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "BATCH_SIZE = 32 #256\n",
    "iteration = 15000 #600000\n",
    "EPOCH = iteration*BATCH_SIZE//int(ds_info.splits['train'].num_examples)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "20"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "EPOCH"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "(ds_train, ds_test), ds_info = tfds.load(\n",
    "    'cats_vs_dogs',\n",
    "    split=['train[:80%]', 'train[80%:]'],\n",
    "    as_supervised=True,\n",
    "    shuffle_files=True,\n",
    "    with_info=True,\n",
    ")\n",
    "\n",
    "ds_train = apply_normalize_on_dataset(ds_train, batch_size=BATCH_SIZE)\n",
    "ds_test = apply_normalize_on_dataset(ds_test, batch_size=BATCH_SIZE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "def scheduler(epoch, lr):\n",
    "    if epoch == EPOCH//2:\n",
    "        lr *=0.1\n",
    "        print(lr)\n",
    "    elif epoch == (EPOCH//4)*3:\n",
    "        lr *=0.1\n",
    "        print(lr)\n",
    "    return lr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "lr_scheduler = keras.callbacks.ReduceLROnPlateau(factor=0.1, monitor='val_loss', patience=1, verbose=2, min_lr=1e-6)\n",
    "# lr_scheduler = keras.callbacks.LearningRateScheduler(scheduler)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "resnet_34 = build_resnet(num_cnn_list = [3, 4, 6, 3],\n",
    "                         channel_list=[64, 128, 256, 512],\n",
    "                         is_50=False,\n",
    "                         cats_vs_dogs=True,\n",
    "                         input_shape=(224,224,3),\n",
    "                         name='ResNet_34')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "654/654 [==============================] - 128s 196ms/step - loss: 2.2930 - accuracy: 0.5697 - val_loss: 2.0546 - val_accuracy: 0.5516\n",
      "Epoch 2/20\n",
      "654/654 [==============================] - ETA: 0s - loss: 1.8510 - accuracy: 0.6281\n",
      "Epoch 00002: ReduceLROnPlateau reducing learning rate to 0.010000000149011612.\n",
      "654/654 [==============================] - 114s 175ms/step - loss: 1.8510 - accuracy: 0.6281 - val_loss: 2.3690 - val_accuracy: 0.5052\n",
      "Epoch 3/20\n",
      "654/654 [==============================] - 114s 175ms/step - loss: 1.6372 - accuracy: 0.6770 - val_loss: 1.6969 - val_accuracy: 0.5938\n",
      "Epoch 4/20\n",
      "654/654 [==============================] - 114s 175ms/step - loss: 1.5478 - accuracy: 0.7342 - val_loss: 1.6013 - val_accuracy: 0.6793\n",
      "Epoch 5/20\n",
      "654/654 [==============================] - ETA: 0s - loss: 1.4663 - accuracy: 0.7687\n",
      "Epoch 00005: ReduceLROnPlateau reducing learning rate to 0.0009999999776482583.\n",
      "654/654 [==============================] - 114s 175ms/step - loss: 1.4663 - accuracy: 0.7687 - val_loss: 1.6579 - val_accuracy: 0.6654\n",
      "Epoch 6/20\n",
      "654/654 [==============================] - 114s 175ms/step - loss: 1.3914 - accuracy: 0.8089 - val_loss: 1.3914 - val_accuracy: 0.8077\n",
      "Epoch 7/20\n",
      "654/654 [==============================] - 115s 175ms/step - loss: 1.3633 - accuracy: 0.8266 - val_loss: 1.3617 - val_accuracy: 0.8207\n",
      "Epoch 8/20\n",
      "654/654 [==============================] - ETA: 0s - loss: 1.3420 - accuracy: 0.8356\n",
      "Epoch 00008: ReduceLROnPlateau reducing learning rate to 9.999999310821295e-05.\n",
      "654/654 [==============================] - 114s 175ms/step - loss: 1.3420 - accuracy: 0.8356 - val_loss: 1.3667 - val_accuracy: 0.8177\n",
      "Epoch 9/20\n",
      "654/654 [==============================] - 114s 175ms/step - loss: 1.3199 - accuracy: 0.8471 - val_loss: 1.3546 - val_accuracy: 0.8312\n",
      "Epoch 10/20\n",
      "654/654 [==============================] - 115s 175ms/step - loss: 1.3199 - accuracy: 0.8455 - val_loss: 1.3501 - val_accuracy: 0.8329\n",
      "Epoch 11/20\n",
      "654/654 [==============================] - 114s 175ms/step - loss: 1.3124 - accuracy: 0.8485 - val_loss: 1.3358 - val_accuracy: 0.8420\n",
      "Epoch 12/20\n",
      "654/654 [==============================] - ETA: 0s - loss: 1.3086 - accuracy: 0.8512\n",
      "Epoch 00012: ReduceLROnPlateau reducing learning rate to 9.999999019782991e-06.\n",
      "654/654 [==============================] - 114s 175ms/step - loss: 1.3086 - accuracy: 0.8512 - val_loss: 1.3445 - val_accuracy: 0.8398\n",
      "Epoch 13/20\n",
      "654/654 [==============================] - 122s 187ms/step - loss: 1.2999 - accuracy: 0.8569 - val_loss: 1.3168 - val_accuracy: 0.8446\n",
      "Epoch 14/20\n",
      "654/654 [==============================] - ETA: 0s - loss: 1.3068 - accuracy: 0.8528\n",
      "Epoch 00014: ReduceLROnPlateau reducing learning rate to 1e-06.\n",
      "654/654 [==============================] - 114s 175ms/step - loss: 1.3068 - accuracy: 0.8528 - val_loss: 1.3393 - val_accuracy: 0.8372\n",
      "Epoch 15/20\n",
      "654/654 [==============================] - 114s 175ms/step - loss: 1.3049 - accuracy: 0.8520 - val_loss: 1.3514 - val_accuracy: 0.8288\n",
      "Epoch 16/20\n",
      "654/654 [==============================] - 115s 175ms/step - loss: 1.3037 - accuracy: 0.8538 - val_loss: 1.3331 - val_accuracy: 0.8416\n",
      "Epoch 17/20\n",
      "654/654 [==============================] - 115s 175ms/step - loss: 1.3053 - accuracy: 0.8514 - val_loss: 1.3387 - val_accuracy: 0.8459\n",
      "Epoch 18/20\n",
      "654/654 [==============================] - 114s 175ms/step - loss: 1.3055 - accuracy: 0.8533 - val_loss: 1.3345 - val_accuracy: 0.8433\n",
      "Epoch 19/20\n",
      "654/654 [==============================] - 115s 175ms/step - loss: 1.3022 - accuracy: 0.8534 - val_loss: 1.3413 - val_accuracy: 0.8407\n",
      "Epoch 20/20\n",
      "654/654 [==============================] - 114s 175ms/step - loss: 1.3026 - accuracy: 0.8521 - val_loss: 1.3308 - val_accuracy: 0.8446\n"
     ]
    }
   ],
   "source": [
    "resnet_34.compile(\n",
    "    loss='binary_crossentropy',\n",
    "    optimizer=tf.keras.optimizers.SGD(lr=0.1, momentum=0.9, clipnorm=1.),\n",
    "    metrics=['accuracy'],\n",
    ")\n",
    "\n",
    "history_34 = resnet_34.fit(\n",
    "    ds_train,\n",
    "    steps_per_epoch=int(ds_info.splits['train[:90%]'].num_examples/BATCH_SIZE),\n",
    "    validation_steps=int(ds_info.splits['train[90%:]'].num_examples/BATCH_SIZE),\n",
    "    epochs=EPOCH,\n",
    "    validation_data=ds_test,\n",
    "    verbose=1,\n",
    "    callbacks=[lr_scheduler],\n",
    "    use_multiprocessing=True,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "resnet_50 = build_resnet(num_cnn_list = [3, 4, 6, 3],\n",
    "                         channel_list=[64, 128, 256, 512],\n",
    "                         is_50=True,\n",
    "                         cats_vs_dogs=True,\n",
    "                         input_shape=(224,224,3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "654/654 [==============================] - 136s 207ms/step - loss: 5.7097 - accuracy: 0.5324 - val_loss: 4.8970 - val_accuracy: 0.5933\n",
      "Epoch 2/20\n",
      "654/654 [==============================] - 82s 125ms/step - loss: 4.3846 - accuracy: 0.6178 - val_loss: 3.8889 - val_accuracy: 0.6397\n",
      "Epoch 3/20\n",
      "654/654 [==============================] - 82s 125ms/step - loss: 3.4830 - accuracy: 0.6685 - val_loss: 3.2149 - val_accuracy: 0.6287\n",
      "Epoch 4/20\n",
      "654/654 [==============================] - 82s 125ms/step - loss: 2.7871 - accuracy: 0.7112 - val_loss: 2.5571 - val_accuracy: 0.6712\n",
      "Epoch 5/20\n",
      "654/654 [==============================] - 82s 125ms/step - loss: 2.2444 - accuracy: 0.7464 - val_loss: 2.2315 - val_accuracy: 0.6064\n",
      "Epoch 6/20\n",
      "654/654 [==============================] - 83s 126ms/step - loss: 1.8377 - accuracy: 0.7682 - val_loss: 1.8668 - val_accuracy: 0.6554\n",
      "Epoch 7/20\n",
      "654/654 [==============================] - 82s 125ms/step - loss: 1.5201 - accuracy: 0.7887 - val_loss: 1.5349 - val_accuracy: 0.6979\n",
      "Epoch 8/20\n",
      "654/654 [==============================] - ETA: 0s - loss: 1.2685 - accuracy: 0.8081\n",
      "Epoch 00008: ReduceLROnPlateau reducing learning rate to 0.010000000149011612.\n",
      "654/654 [==============================] - 82s 125ms/step - loss: 1.2685 - accuracy: 0.8081 - val_loss: 1.9616 - val_accuracy: 0.5543\n",
      "Epoch 9/20\n",
      "654/654 [==============================] - 83s 128ms/step - loss: 1.0444 - accuracy: 0.8736 - val_loss: 1.0631 - val_accuracy: 0.8602\n",
      "Epoch 10/20\n",
      "654/654 [==============================] - ETA: 0s - loss: 0.9671 - accuracy: 0.9023\n",
      "Epoch 00010: ReduceLROnPlateau reducing learning rate to 0.0009999999776482583.\n",
      "654/654 [==============================] - 84s 128ms/step - loss: 0.9671 - accuracy: 0.9023 - val_loss: 1.2204 - val_accuracy: 0.8148\n",
      "Epoch 11/20\n",
      "654/654 [==============================] - 82s 125ms/step - loss: 0.9169 - accuracy: 0.9205 - val_loss: 1.0425 - val_accuracy: 0.8707\n",
      "Epoch 12/20\n",
      "654/654 [==============================] - 83s 127ms/step - loss: 0.8955 - accuracy: 0.9291 - val_loss: 1.0369 - val_accuracy: 0.8772\n",
      "Epoch 13/20\n",
      "654/654 [==============================] - ETA: 0s - loss: 0.8872 - accuracy: 0.9321\n",
      "Epoch 00013: ReduceLROnPlateau reducing learning rate to 9.999999310821295e-05.\n",
      "654/654 [==============================] - 82s 126ms/step - loss: 0.8872 - accuracy: 0.9321 - val_loss: 1.0612 - val_accuracy: 0.8724\n",
      "Epoch 14/20\n",
      "654/654 [==============================] - ETA: 0s - loss: 0.8796 - accuracy: 0.9372\n",
      "Epoch 00014: ReduceLROnPlateau reducing learning rate to 9.999999019782991e-06.\n",
      "654/654 [==============================] - 82s 126ms/step - loss: 0.8796 - accuracy: 0.9372 - val_loss: 1.0556 - val_accuracy: 0.8715\n",
      "Epoch 15/20\n",
      "654/654 [==============================] - ETA: 0s - loss: 0.8778 - accuracy: 0.9350\n",
      "Epoch 00015: ReduceLROnPlateau reducing learning rate to 1e-06.\n",
      "654/654 [==============================] - 82s 126ms/step - loss: 0.8778 - accuracy: 0.9350 - val_loss: 1.0375 - val_accuracy: 0.8739\n",
      "Epoch 16/20\n",
      "654/654 [==============================] - 82s 126ms/step - loss: 0.8758 - accuracy: 0.9367 - val_loss: 1.0760 - val_accuracy: 0.8620\n",
      "Epoch 17/20\n",
      "654/654 [==============================] - 82s 126ms/step - loss: 0.8735 - accuracy: 0.9398 - val_loss: 1.0288 - val_accuracy: 0.8824\n",
      "Epoch 18/20\n",
      "654/654 [==============================] - 82s 126ms/step - loss: 0.8749 - accuracy: 0.9379 - val_loss: 1.0356 - val_accuracy: 0.8733\n",
      "Epoch 19/20\n",
      "654/654 [==============================] - 82s 126ms/step - loss: 0.8757 - accuracy: 0.9362 - val_loss: 1.0691 - val_accuracy: 0.8665\n",
      "Epoch 20/20\n",
      "654/654 [==============================] - 83s 127ms/step - loss: 0.8762 - accuracy: 0.9362 - val_loss: 1.0725 - val_accuracy: 0.8676\n"
     ]
    }
   ],
   "source": [
    "resnet_50.compile(\n",
    "    loss='binary_crossentropy',\n",
    "    optimizer=tf.keras.optimizers.SGD(lr=0.1, momentum=0.9, clipnorm=1.),\n",
    "    metrics=['accuracy'],\n",
    ")\n",
    "\n",
    "history_50 = resnet_50.fit(\n",
    "    ds_train,\n",
    "    steps_per_epoch=int(ds_info.splits['train[:90%]'].num_examples/BATCH_SIZE),\n",
    "    validation_steps=int(ds_info.splits['train[90%:]'].num_examples/BATCH_SIZE),\n",
    "    epochs=EPOCH,\n",
    "    validation_data=ds_test,\n",
    "    verbose=1,\n",
    "    callbacks=[lr_scheduler],\n",
    "    use_multiprocessing=True,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "plain_34 = build_plain(num_cnn_list=[3, 4, 6, 3],\n",
    "                       channel_list=[64, 128, 256, 512],\n",
    "                       is_50=False,\n",
    "                       cats_vs_dogs=True,\n",
    "                       input_shape=(224,224,3),\n",
    "                       name='Plain_34')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "654/654 [==============================] - 109s 166ms/step - loss: 2.0811 - accuracy: 0.5047 - val_loss: 1.8983 - val_accuracy: 0.5123\n",
      "Epoch 2/20\n",
      "654/654 [==============================] - 108s 165ms/step - loss: 1.7524 - accuracy: 0.5103 - val_loss: 1.6206 - val_accuracy: 0.4939\n",
      "Epoch 3/20\n",
      "654/654 [==============================] - 109s 167ms/step - loss: 1.5037 - accuracy: 0.5284 - val_loss: 1.4057 - val_accuracy: 0.5206\n",
      "Epoch 4/20\n",
      "654/654 [==============================] - ETA: 0s - loss: 1.3063 - accuracy: 0.5629\n",
      "Epoch 00004: ReduceLROnPlateau reducing learning rate to 0.010000000149011612.\n",
      "654/654 [==============================] - 109s 166ms/step - loss: 1.3063 - accuracy: 0.5629 - val_loss: 1.4876 - val_accuracy: 0.5330\n",
      "Epoch 5/20\n",
      "654/654 [==============================] - 109s 166ms/step - loss: 1.2031 - accuracy: 0.5901 - val_loss: 1.1892 - val_accuracy: 0.5963\n",
      "Epoch 6/20\n",
      "654/654 [==============================] - 109s 167ms/step - loss: 1.1732 - accuracy: 0.6192 - val_loss: 1.1762 - val_accuracy: 0.5881\n",
      "Epoch 7/20\n",
      "654/654 [==============================] - 109s 166ms/step - loss: 1.1442 - accuracy: 0.6416 - val_loss: 1.1543 - val_accuracy: 0.6024\n",
      "Epoch 8/20\n",
      "654/654 [==============================] - ETA: 0s - loss: 1.1239 - accuracy: 0.6442\n",
      "Epoch 00008: ReduceLROnPlateau reducing learning rate to 0.0009999999776482583.\n",
      "654/654 [==============================] - 109s 166ms/step - loss: 1.1239 - accuracy: 0.6442 - val_loss: 1.3338 - val_accuracy: 0.4991\n",
      "Epoch 9/20\n",
      "654/654 [==============================] - 109s 167ms/step - loss: 1.0977 - accuracy: 0.6679 - val_loss: 1.1122 - val_accuracy: 0.6489\n",
      "Epoch 10/20\n",
      "654/654 [==============================] - 109s 166ms/step - loss: 1.0831 - accuracy: 0.6763 - val_loss: 1.0764 - val_accuracy: 0.6979\n",
      "Epoch 11/20\n",
      "654/654 [==============================] - ETA: 0s - loss: 1.0776 - accuracy: 0.6812\n",
      "Epoch 00011: ReduceLROnPlateau reducing learning rate to 9.999999310821295e-05.\n",
      "654/654 [==============================] - 109s 166ms/step - loss: 1.0776 - accuracy: 0.6812 - val_loss: 1.1071 - val_accuracy: 0.6532\n",
      "Epoch 12/20\n",
      "654/654 [==============================] - 109s 167ms/step - loss: 1.0654 - accuracy: 0.6909 - val_loss: 1.0569 - val_accuracy: 0.6966\n",
      "Epoch 13/20\n",
      "654/654 [==============================] - 109s 167ms/step - loss: 1.0648 - accuracy: 0.6934 - val_loss: 1.0532 - val_accuracy: 0.6997\n",
      "Epoch 14/20\n",
      "654/654 [==============================] - 109s 167ms/step - loss: 1.0613 - accuracy: 0.6946 - val_loss: 1.0445 - val_accuracy: 0.7083\n",
      "Epoch 15/20\n",
      "654/654 [==============================] - ETA: 0s - loss: 1.0605 - accuracy: 0.6968\n",
      "Epoch 00015: ReduceLROnPlateau reducing learning rate to 9.999999019782991e-06.\n",
      "654/654 [==============================] - 109s 167ms/step - loss: 1.0605 - accuracy: 0.6968 - val_loss: 1.0620 - val_accuracy: 0.6862\n",
      "Epoch 16/20\n",
      "654/654 [==============================] - ETA: 0s - loss: 1.0566 - accuracy: 0.6979\n",
      "Epoch 00016: ReduceLROnPlateau reducing learning rate to 1e-06.\n",
      "654/654 [==============================] - 109s 167ms/step - loss: 1.0566 - accuracy: 0.6979 - val_loss: 1.0619 - val_accuracy: 0.6896\n",
      "Epoch 17/20\n",
      "654/654 [==============================] - 109s 167ms/step - loss: 1.0581 - accuracy: 0.6987 - val_loss: 1.0587 - val_accuracy: 0.6897\n",
      "Epoch 18/20\n",
      "654/654 [==============================] - 109s 167ms/step - loss: 1.0615 - accuracy: 0.6988 - val_loss: 1.0491 - val_accuracy: 0.6979\n",
      "Epoch 19/20\n",
      "654/654 [==============================] - 109s 166ms/step - loss: 1.0572 - accuracy: 0.6989 - val_loss: 1.0540 - val_accuracy: 0.6910\n",
      "Epoch 20/20\n",
      "654/654 [==============================] - 109s 167ms/step - loss: 1.0559 - accuracy: 0.6997 - val_loss: 1.0483 - val_accuracy: 0.6949\n"
     ]
    }
   ],
   "source": [
    "plain_34.compile(\n",
    "    loss='binary_crossentropy',\n",
    "    optimizer=tf.keras.optimizers.SGD(lr=0.1, momentum=0.9, clipnorm=1.),\n",
    "    metrics=['accuracy'],\n",
    ")\n",
    "\n",
    "history_plain_34 = plain_34.fit(\n",
    "    ds_train,\n",
    "    steps_per_epoch=int(ds_info.splits['train[:90%]'].num_examples/BATCH_SIZE),\n",
    "    validation_steps=int(ds_info.splits['train[90%:]'].num_examples/BATCH_SIZE),\n",
    "    epochs=EPOCH,\n",
    "    validation_data=ds_test,\n",
    "    verbose=1,\n",
    "    callbacks=[lr_scheduler],\n",
    "    use_multiprocessing=True,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "plain_50 = build_plain(num_cnn_list = [3, 4, 6, 3],\n",
    "                       channel_list=[64, 128, 256, 512],\n",
    "                       is_50=True,\n",
    "                       cats_vs_dogs=True,\n",
    "                       input_shape=(224,224,3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "654/654 [==============================] - 201s 307ms/step - loss: 4.8451 - accuracy: 0.5014 - val_loss: 4.2858 - val_accuracy: 0.5165\n",
      "Epoch 2/20\n",
      "654/654 [==============================] - 171s 262ms/step - loss: 3.8565 - accuracy: 0.5030 - val_loss: 3.4590 - val_accuracy: 0.5065\n",
      "Epoch 3/20\n",
      "654/654 [==============================] - 183s 279ms/step - loss: 3.1262 - accuracy: 0.5079 - val_loss: 2.8161 - val_accuracy: 0.4969\n",
      "Epoch 4/20\n",
      "654/654 [==============================] - 171s 262ms/step - loss: 2.5623 - accuracy: 0.5447 - val_loss: 2.3260 - val_accuracy: 0.5534\n",
      "Epoch 5/20\n",
      "654/654 [==============================] - 172s 262ms/step - loss: 2.1252 - accuracy: 0.5720 - val_loss: 1.9800 - val_accuracy: 0.4905\n",
      "Epoch 6/20\n",
      "654/654 [==============================] - 172s 262ms/step - loss: 1.7962 - accuracy: 0.5685 - val_loss: 1.7431 - val_accuracy: 0.5136\n",
      "Epoch 7/20\n",
      "654/654 [==============================] - 172s 263ms/step - loss: 1.5356 - accuracy: 0.5865 - val_loss: 1.4828 - val_accuracy: 0.5612\n",
      "Epoch 8/20\n",
      "654/654 [==============================] - 172s 262ms/step - loss: 1.3296 - accuracy: 0.6005 - val_loss: 1.2589 - val_accuracy: 0.5846\n",
      "Epoch 9/20\n",
      "654/654 [==============================] - 172s 262ms/step - loss: 1.1682 - accuracy: 0.6180 - val_loss: 1.1313 - val_accuracy: 0.6095\n",
      "Epoch 10/20\n",
      "654/654 [==============================] - 172s 263ms/step - loss: 1.0767 - accuracy: 0.5822 - val_loss: 1.0466 - val_accuracy: 0.5661\n",
      "Epoch 11/20\n",
      "654/654 [==============================] - 171s 262ms/step - loss: 0.9601 - accuracy: 0.6402 - val_loss: 0.9680 - val_accuracy: 0.5694\n",
      "Epoch 12/20\n",
      "654/654 [==============================] - 172s 263ms/step - loss: 0.8943 - accuracy: 0.6358 - val_loss: 0.8913 - val_accuracy: 0.6480\n",
      "Epoch 13/20\n",
      "654/654 [==============================] - 172s 262ms/step - loss: 0.8286 - accuracy: 0.6566 - val_loss: 0.7999 - val_accuracy: 0.6699\n",
      "Epoch 14/20\n",
      "654/654 [==============================] - ETA: 0s - loss: 0.7755 - accuracy: 0.6758\n",
      "Epoch 00014: ReduceLROnPlateau reducing learning rate to 0.010000000149011612.\n",
      "654/654 [==============================] - 172s 263ms/step - loss: 0.7755 - accuracy: 0.6758 - val_loss: 0.8247 - val_accuracy: 0.6337\n",
      "Epoch 15/20\n",
      "654/654 [==============================] - 172s 263ms/step - loss: 0.7296 - accuracy: 0.6979 - val_loss: 0.7290 - val_accuracy: 0.6866\n",
      "Epoch 16/20\n",
      "654/654 [==============================] - 172s 263ms/step - loss: 0.7117 - accuracy: 0.7070 - val_loss: 0.7264 - val_accuracy: 0.6935\n",
      "Epoch 17/20\n",
      "654/654 [==============================] - 172s 263ms/step - loss: 0.7032 - accuracy: 0.7139 - val_loss: 0.7085 - val_accuracy: 0.7127\n",
      "Epoch 18/20\n",
      "654/654 [==============================] - 172s 263ms/step - loss: 0.6880 - accuracy: 0.7204 - val_loss: 0.6898 - val_accuracy: 0.7229\n",
      "Epoch 19/20\n",
      "654/654 [==============================] - ETA: 0s - loss: 0.6779 - accuracy: 0.7245\n",
      "Epoch 00019: ReduceLROnPlateau reducing learning rate to 0.0009999999776482583.\n",
      "654/654 [==============================] - 172s 263ms/step - loss: 0.6779 - accuracy: 0.7245 - val_loss: 0.7373 - val_accuracy: 0.7057\n",
      "Epoch 20/20\n",
      "654/654 [==============================] - 172s 263ms/step - loss: 0.6643 - accuracy: 0.7347 - val_loss: 0.6766 - val_accuracy: 0.7292\n"
     ]
    }
   ],
   "source": [
    "plain_50.compile(\n",
    "    loss='binary_crossentropy',\n",
    "    optimizer=tf.keras.optimizers.SGD(lr=0.1, momentum=0.9, clipnorm=1.),\n",
    "    metrics=['accuracy'],\n",
    ")\n",
    "\n",
    "history_plain_50 = plain_50.fit(\n",
    "    ds_train,\n",
    "    steps_per_epoch=int(ds_info.splits['train[:90%]'].num_examples/BATCH_SIZE),\n",
    "    validation_steps=int(ds_info.splits['train[90%:]'].num_examples/BATCH_SIZE),\n",
    "    epochs=EPOCH,\n",
    "    validation_data=ds_test,\n",
    "    verbose=1,\n",
    "    callbacks=[lr_scheduler],\n",
    "    use_multiprocessing=True,\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Comparison"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history_34.history['val_accuracy'], 'b')\n",
    "plt.plot(history_50.history['val_accuracy'], 'g')\n",
    "plt.plot(history_plain_34.history['val_accuracy'], 'r')\n",
    "plt.plot(history_plain_50.history['val_accuracy'], 'm')\n",
    "plt.title('Model validation accuracy')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['resnet_34', 'resnet_50', 'plain_34', 'plain_50'], loc='lower right')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Report"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "ResNet-34 와 ResNet-50을 구현하여, Plain-34, Plain-50과 비교해봤습니다.  \n",
    "데이터는 cats_vs_dogs 데이터를 사용했습니다."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "aiffel",
   "language": "python",
   "name": "aiffel"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": true,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {
    "height": "calc(100% - 180px)",
    "left": "10px",
    "top": "150px",
    "width": "384px"
   },
   "toc_section_display": true,
   "toc_window_display": true
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}