custom_convnets_exercise.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "19023a3b-8d7e-4736-8381-9eed66c488d4",
   "metadata": {},
   "source": [
    "# Introduction #\n",
    "\n",
    "In these exercises, you'll build a custom convnet with performance competitive to the VGG16 model from Lesson 1.\n",
    "\n",
    "Get started by running the code cell below."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "b58ea111-8d9f-4682-a1be-1bee07047a2c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 5117 files belonging to 2 classes.\n",
      "Found 5051 files belonging to 2 classes.\n"
     ]
    }
   ],
   "source": [
    "# Imports\n",
    "import os, warnings\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib import gridspec\n",
    "\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras.preprocessing import image_dataset_from_directory\n",
    "\n",
    "# Reproducability\n",
    "def set_seed(seed=31415):\n",
    "    np.random.seed(seed)\n",
    "    tf.random.set_seed(seed)\n",
    "    os.environ['PYTHONHASHSEED'] = str(seed)\n",
    "    os.environ['TF_DETERMINISTIC_OPS'] = '1'\n",
    "set_seed()\n",
    "\n",
    "# Set Matplotlib defaults\n",
    "plt.rc('figure', autolayout=True)\n",
    "plt.rc('axes', labelweight='bold', labelsize='large',\n",
    "       titleweight='bold', titlesize=18, titlepad=10)\n",
    "plt.rc('image', cmap='magma')\n",
    "warnings.filterwarnings(\"ignore\") # to clean up output cells\n",
    "\n",
    "\n",
    "# Load training and validation sets\n",
    "ds_train_ = image_dataset_from_directory(\n",
    "    './input/car-or-truck/train',\n",
    "    labels='inferred',\n",
    "    label_mode='binary',\n",
    "    image_size=[128, 128],\n",
    "    interpolation='nearest',\n",
    "    batch_size=64,\n",
    "    shuffle=True,\n",
    ")\n",
    "ds_valid_ = image_dataset_from_directory(\n",
    "    './input/car-or-truck/valid',\n",
    "    labels='inferred',\n",
    "    label_mode='binary',\n",
    "    image_size=[128, 128],\n",
    "    interpolation='nearest',\n",
    "    batch_size=64,\n",
    "    shuffle=False,\n",
    ")\n",
    "\n",
    "# Data Pipeline\n",
    "def convert_to_float(image, label):\n",
    "    image = tf.image.convert_image_dtype(image, dtype=tf.float32)\n",
    "    return image, label\n",
    "\n",
    "AUTOTUNE = tf.data.experimental.AUTOTUNE\n",
    "ds_train = (\n",
    "    ds_train_\n",
    "    .map(convert_to_float)\n",
    "    .cache()\n",
    "    .prefetch(buffer_size=AUTOTUNE)\n",
    ")\n",
    "ds_valid = (\n",
    "    ds_valid_\n",
    "    .map(convert_to_float)\n",
    "    .cache()\n",
    "    .prefetch(buffer_size=AUTOTUNE)\n",
    ")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "58584b19-0d9a-4451-b1c7-85a8330a4103",
   "metadata": {},
   "source": [
    "# Design a Convnet #\n",
    "\n",
    "Let's design a convolutional network with a block architecture like we saw in the tutorial. The model from the example had three blocks, each with a single convolutional layer. Its performance on the \"Car or Truck\" problem was okay, but far from what the pretrained VGG16 could achieve. It might be that our simple network lacks the ability to extract sufficiently complex features. We could try improving the model either by adding more blocks or by adding convolutions to the blocks we have.\n",
    "\n",
    "Let's go with the second approach. We'll keep the three block structure, but increase the number of `Conv2D` layer in the second block to two, and in the third block to three.\n",
    "\n",
    "<figure>\n",
    "<!-- <img src=\"./images/2-convmodel-2.png\" width=\"250\" alt=\"Diagram of a convolutional model.\"> -->\n",
    "<img src=\"refimages/cnn023.png\" width=\"250\" alt=\"Diagram of a convolutional model.\">\n",
    "</figure>\n",
    "\n",
    "# 1) Define Model #\n",
    "\n",
    "Given the diagram above, complete the model by defining the layers of the third block."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "2eaf417d-dc7d-4b4d-9fae-5c9afe78e63a",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow import keras\n",
    "from tensorflow.keras import layers\n",
    "\n",
    "model = keras.Sequential([\n",
    "    # Block One\n",
    "    layers.Conv2D(filters=32, kernel_size=3, activation='relu', padding='same',\n",
    "                  input_shape=[128, 128, 3]),\n",
    "    layers.MaxPool2D(),\n",
    "\n",
    "    # Block Two\n",
    "    layers.Conv2D(filters=64, kernel_size=3, activation='relu', padding='same'),\n",
    "    layers.MaxPool2D(),\n",
    "    \n",
    "    # Block Three\n",
    "    layers.Conv2D(filters=128, kernel_size=3, activation='relu', padding='same'),\n",
    "    layers.Conv2D(filters=128, kernel_size=3, activation='relu', padding='same'),\n",
    "    layers.MaxPool2D(),\n",
    "\n",
    "    # Head\n",
    "    layers.Flatten(),\n",
    "    layers.Dense(6, activation='relu'),\n",
    "    layers.Dropout(0.2),\n",
    "    layers.Dense(1, activation='sigmoid'),\n",
    "])\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "759eb71d-bb91-4dab-bebf-2023639afcc4",
   "metadata": {},
   "source": [
    "# 2) Compile #\n",
    "\n",
    "To prepare for training, compile the model with an appropriate loss and accuracy metric for the \"Car or Truck\" dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "e5d48558-3e07-4a9e-acb0-4902818e6e09",
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(\n",
    "    optimizer=tf.keras.optimizers.Adam(epsilon=0.01),\n",
    "    # YOUR CODE HERE: Add loss and metric\n",
    "     loss='binary_crossentropy',\n",
    "    metrics=['binary_accuracy'],\n",
    ")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e59287b4-728c-4169-ade2-767cb8bcd6b7",
   "metadata": {},
   "source": [
    "Finally, let's test the performance of this new model. First run this cell to fit the model to the training set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "432f86e6-ba61-4e2f-938f-4a8cfbf18846",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "80/80 [==============================] - 318s 4s/step - loss: 0.6921 - binary_accuracy: 0.5550 - val_loss: 0.6903 - val_binary_accuracy: 0.5785\n",
      "Epoch 2/50\n",
      "80/80 [==============================] - 237s 3s/step - loss: 0.6888 - binary_accuracy: 0.5806 - val_loss: 0.6876 - val_binary_accuracy: 0.5785\n",
      "Epoch 3/50\n",
      "34/80 [===========>..................] - ETA: 1:27 - loss: 0.6879 - binary_accuracy: 0.5699"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[5], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m history \u001b[38;5;241m=\u001b[39m \u001b[43mmodel\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfit\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m      2\u001b[0m \u001b[43m    \u001b[49m\u001b[43mds_train\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m      3\u001b[0m \u001b[43m    \u001b[49m\u001b[43mvalidation_data\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mds_valid\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m      4\u001b[0m \u001b[43m    \u001b[49m\u001b[43mepochs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m50\u001b[39;49m\u001b[43m,\u001b[49m\n\u001b[1;32m      5\u001b[0m \u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m~/projects/opencomp_vision/env/lib/python3.10/site-packages/keras/src/utils/traceback_utils.py:65\u001b[0m, in \u001b[0;36mfilter_traceback.<locals>.error_handler\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m     63\u001b[0m filtered_tb \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[1;32m     64\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m---> 65\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mfn\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     66\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mException\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[1;32m     67\u001b[0m     filtered_tb \u001b[38;5;241m=\u001b[39m _process_traceback_frames(e\u001b[38;5;241m.\u001b[39m__traceback__)\n",
      "File \u001b[0;32m~/projects/opencomp_vision/env/lib/python3.10/site-packages/keras/src/engine/training.py:1742\u001b[0m, in \u001b[0;36mModel.fit\u001b[0;34m(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_batch_size, validation_freq, max_queue_size, workers, use_multiprocessing)\u001b[0m\n\u001b[1;32m   1734\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m tf\u001b[38;5;241m.\u001b[39mprofiler\u001b[38;5;241m.\u001b[39mexperimental\u001b[38;5;241m.\u001b[39mTrace(\n\u001b[1;32m   1735\u001b[0m     \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mtrain\u001b[39m\u001b[38;5;124m\"\u001b[39m,\n\u001b[1;32m   1736\u001b[0m     epoch_num\u001b[38;5;241m=\u001b[39mepoch,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m   1739\u001b[0m     _r\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m1\u001b[39m,\n\u001b[1;32m   1740\u001b[0m ):\n\u001b[1;32m   1741\u001b[0m     callbacks\u001b[38;5;241m.\u001b[39mon_train_batch_begin(step)\n\u001b[0;32m-> 1742\u001b[0m     tmp_logs \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mtrain_function\u001b[49m\u001b[43m(\u001b[49m\u001b[43miterator\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1743\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m data_handler\u001b[38;5;241m.\u001b[39mshould_sync:\n\u001b[1;32m   1744\u001b[0m         context\u001b[38;5;241m.\u001b[39masync_wait()\n",
      "File \u001b[0;32m~/projects/opencomp_vision/env/lib/python3.10/site-packages/tensorflow/python/util/traceback_utils.py:150\u001b[0m, in \u001b[0;36mfilter_traceback.<locals>.error_handler\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m    148\u001b[0m filtered_tb \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[1;32m    149\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m--> 150\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mfn\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    151\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mException\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[1;32m    152\u001b[0m   filtered_tb \u001b[38;5;241m=\u001b[39m _process_traceback_frames(e\u001b[38;5;241m.\u001b[39m__traceback__)\n",
      "File \u001b[0;32m~/projects/opencomp_vision/env/lib/python3.10/site-packages/tensorflow/python/eager/polymorphic_function/polymorphic_function.py:825\u001b[0m, in \u001b[0;36mFunction.__call__\u001b[0;34m(self, *args, **kwds)\u001b[0m\n\u001b[1;32m    822\u001b[0m compiler \u001b[38;5;241m=\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mxla\u001b[39m\u001b[38;5;124m\"\u001b[39m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_jit_compile \u001b[38;5;28;01melse\u001b[39;00m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mnonXla\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m    824\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m OptionalXlaContext(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_jit_compile):\n\u001b[0;32m--> 825\u001b[0m   result \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_call\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwds\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    827\u001b[0m new_tracing_count \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mexperimental_get_tracing_count()\n\u001b[1;32m    828\u001b[0m without_tracing \u001b[38;5;241m=\u001b[39m (tracing_count \u001b[38;5;241m==\u001b[39m new_tracing_count)\n",
      "File \u001b[0;32m~/projects/opencomp_vision/env/lib/python3.10/site-packages/tensorflow/python/eager/polymorphic_function/polymorphic_function.py:857\u001b[0m, in \u001b[0;36mFunction._call\u001b[0;34m(self, *args, **kwds)\u001b[0m\n\u001b[1;32m    854\u001b[0m   \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_lock\u001b[38;5;241m.\u001b[39mrelease()\n\u001b[1;32m    855\u001b[0m   \u001b[38;5;66;03m# In this case we have created variables on the first call, so we run the\u001b[39;00m\n\u001b[1;32m    856\u001b[0m   \u001b[38;5;66;03m# defunned version which is guaranteed to never create variables.\u001b[39;00m\n\u001b[0;32m--> 857\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_no_variable_creation_fn\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwds\u001b[49m\u001b[43m)\u001b[49m  \u001b[38;5;66;03m# pylint: disable=not-callable\u001b[39;00m\n\u001b[1;32m    858\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_variable_creation_fn \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[1;32m    859\u001b[0m   \u001b[38;5;66;03m# Release the lock early so that multiple threads can perform the call\u001b[39;00m\n\u001b[1;32m    860\u001b[0m   \u001b[38;5;66;03m# in parallel.\u001b[39;00m\n\u001b[1;32m    861\u001b[0m   \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_lock\u001b[38;5;241m.\u001b[39mrelease()\n",
      "File \u001b[0;32m~/projects/opencomp_vision/env/lib/python3.10/site-packages/tensorflow/python/eager/polymorphic_function/tracing_compiler.py:148\u001b[0m, in \u001b[0;36mTracingCompiler.__call__\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m    145\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_lock:\n\u001b[1;32m    146\u001b[0m   (concrete_function,\n\u001b[1;32m    147\u001b[0m    filtered_flat_args) \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_maybe_define_function(args, kwargs)\n\u001b[0;32m--> 148\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mconcrete_function\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_call_flat\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    149\u001b[0m \u001b[43m    \u001b[49m\u001b[43mfiltered_flat_args\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcaptured_inputs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mconcrete_function\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mcaptured_inputs\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m~/projects/opencomp_vision/env/lib/python3.10/site-packages/tensorflow/python/eager/polymorphic_function/monomorphic_function.py:1349\u001b[0m, in \u001b[0;36mConcreteFunction._call_flat\u001b[0;34m(self, args, captured_inputs)\u001b[0m\n\u001b[1;32m   1345\u001b[0m possible_gradient_type \u001b[38;5;241m=\u001b[39m gradients_util\u001b[38;5;241m.\u001b[39mPossibleTapeGradientTypes(args)\n\u001b[1;32m   1346\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m (possible_gradient_type \u001b[38;5;241m==\u001b[39m gradients_util\u001b[38;5;241m.\u001b[39mPOSSIBLE_GRADIENT_TYPES_NONE\n\u001b[1;32m   1347\u001b[0m     \u001b[38;5;129;01mand\u001b[39;00m executing_eagerly):\n\u001b[1;32m   1348\u001b[0m   \u001b[38;5;66;03m# No tape is watching; skip to running the function.\u001b[39;00m\n\u001b[0;32m-> 1349\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_build_call_outputs(\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_inference_function\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m)\u001b[49m)\n\u001b[1;32m   1350\u001b[0m forward_backward \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_select_forward_and_backward_functions(\n\u001b[1;32m   1351\u001b[0m     args,\n\u001b[1;32m   1352\u001b[0m     possible_gradient_type,\n\u001b[1;32m   1353\u001b[0m     executing_eagerly)\n\u001b[1;32m   1354\u001b[0m forward_function, args_with_tangents \u001b[38;5;241m=\u001b[39m forward_backward\u001b[38;5;241m.\u001b[39mforward()\n",
      "File \u001b[0;32m~/projects/opencomp_vision/env/lib/python3.10/site-packages/tensorflow/python/eager/polymorphic_function/atomic_function.py:196\u001b[0m, in \u001b[0;36mAtomicFunction.__call__\u001b[0;34m(self, *args)\u001b[0m\n\u001b[1;32m    194\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m record\u001b[38;5;241m.\u001b[39mstop_recording():\n\u001b[1;32m    195\u001b[0m   \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_bound_context\u001b[38;5;241m.\u001b[39mexecuting_eagerly():\n\u001b[0;32m--> 196\u001b[0m     outputs \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_bound_context\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mcall_function\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    197\u001b[0m \u001b[43m        \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mname\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    198\u001b[0m \u001b[43m        \u001b[49m\u001b[38;5;28;43mlist\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43margs\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    199\u001b[0m \u001b[43m        \u001b[49m\u001b[38;5;28;43mlen\u001b[39;49m\u001b[43m(\u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfunction_type\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mflat_outputs\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    200\u001b[0m \u001b[43m    \u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    201\u001b[0m   \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m    202\u001b[0m     outputs \u001b[38;5;241m=\u001b[39m make_call_op_in_graph(\u001b[38;5;28mself\u001b[39m, \u001b[38;5;28mlist\u001b[39m(args))\n",
      "File \u001b[0;32m~/projects/opencomp_vision/env/lib/python3.10/site-packages/tensorflow/python/eager/context.py:1457\u001b[0m, in \u001b[0;36mContext.call_function\u001b[0;34m(self, name, tensor_inputs, num_outputs)\u001b[0m\n\u001b[1;32m   1455\u001b[0m cancellation_context \u001b[38;5;241m=\u001b[39m cancellation\u001b[38;5;241m.\u001b[39mcontext()\n\u001b[1;32m   1456\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m cancellation_context \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[0;32m-> 1457\u001b[0m   outputs \u001b[38;5;241m=\u001b[39m \u001b[43mexecute\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mexecute\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m   1458\u001b[0m \u001b[43m      \u001b[49m\u001b[43mname\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mdecode\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mutf-8\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1459\u001b[0m \u001b[43m      \u001b[49m\u001b[43mnum_outputs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mnum_outputs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1460\u001b[0m \u001b[43m      \u001b[49m\u001b[43minputs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mtensor_inputs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1461\u001b[0m \u001b[43m      \u001b[49m\u001b[43mattrs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mattrs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1462\u001b[0m \u001b[43m      \u001b[49m\u001b[43mctx\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1463\u001b[0m \u001b[43m  \u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1464\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m   1465\u001b[0m   outputs \u001b[38;5;241m=\u001b[39m execute\u001b[38;5;241m.\u001b[39mexecute_with_cancellation(\n\u001b[1;32m   1466\u001b[0m       name\u001b[38;5;241m.\u001b[39mdecode(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mutf-8\u001b[39m\u001b[38;5;124m\"\u001b[39m),\n\u001b[1;32m   1467\u001b[0m       num_outputs\u001b[38;5;241m=\u001b[39mnum_outputs,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m   1471\u001b[0m       cancellation_manager\u001b[38;5;241m=\u001b[39mcancellation_context,\n\u001b[1;32m   1472\u001b[0m   )\n",
      "File \u001b[0;32m~/projects/opencomp_vision/env/lib/python3.10/site-packages/tensorflow/python/eager/execute.py:53\u001b[0m, in \u001b[0;36mquick_execute\u001b[0;34m(op_name, num_outputs, inputs, attrs, ctx, name)\u001b[0m\n\u001b[1;32m     51\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[1;32m     52\u001b[0m   ctx\u001b[38;5;241m.\u001b[39mensure_initialized()\n\u001b[0;32m---> 53\u001b[0m   tensors \u001b[38;5;241m=\u001b[39m \u001b[43mpywrap_tfe\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mTFE_Py_Execute\u001b[49m\u001b[43m(\u001b[49m\u001b[43mctx\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_handle\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdevice_name\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mop_name\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m     54\u001b[0m \u001b[43m                                      \u001b[49m\u001b[43minputs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mattrs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mnum_outputs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     55\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m core\u001b[38;5;241m.\u001b[39m_NotOkStatusException \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[1;32m     56\u001b[0m   \u001b[38;5;28;01mif\u001b[39;00m name \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "history = model.fit(\n",
    "    ds_train,\n",
    "    validation_data=ds_valid,\n",
    "    epochs=50,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "25e6be44-5f16-4e15-884d-0a1fb3a9b17b",
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'history' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[6], line 2\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mpandas\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m \u001b[38;5;21;01mpd\u001b[39;00m\n\u001b[0;32m----> 2\u001b[0m history_frame \u001b[38;5;241m=\u001b[39m pd\u001b[38;5;241m.\u001b[39mDataFrame(\u001b[43mhistory\u001b[49m\u001b[38;5;241m.\u001b[39mhistory)\n\u001b[1;32m      3\u001b[0m history_frame\u001b[38;5;241m.\u001b[39mloc[:, [\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mloss\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mval_loss\u001b[39m\u001b[38;5;124m'\u001b[39m]]\u001b[38;5;241m.\u001b[39mplot()\n\u001b[1;32m      4\u001b[0m history_frame\u001b[38;5;241m.\u001b[39mloc[:, [\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mbinary_accuracy\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mval_binary_accuracy\u001b[39m\u001b[38;5;124m'\u001b[39m]]\u001b[38;5;241m.\u001b[39mplot();\n",
      "\u001b[0;31mNameError\u001b[0m: name 'history' is not defined"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "history_frame = pd.DataFrame(history.history)\n",
    "history_frame.loc[:, ['loss', 'val_loss']].plot()\n",
    "history_frame.loc[:, ['binary_accuracy', 'val_binary_accuracy']].plot();"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "210e3cae-2def-4dde-8b53-e6011c5bfd55",
   "metadata": {},
   "source": [
    "# 3) Train the Model\n",
    "\n",
    "How would you interpret these training curves? Did this model improve upon the model from the tutorial?\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d8325e2f-4f12-43d9-ae59-1aa469f7db2f",
   "metadata": {},
   "source": [
    "The learning curves for the model from the tutorial diverged fairly rapidly. This would indicate that it was prone to overfitting and in need of some regularization. The additional layer in our new model would make it even more prone to overfitting. However, adding some regularization with the Dropout layer helped prevent this. These changes improved the validation accuracy of the model by several points."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b8315c26-d7fc-4245-9e32-cc00bdd92684",
   "metadata": {},
   "source": [
    "# Conclusion\n",
    "\n",
    "These exercises showed you how to design a custom convolutional network to solve a specific classification problem. Though most models these days will be built on top of a pretrained base, it certain circumstances a smaller custom convnet might still be preferable -- such as with a smaller or unusual dataset or when computing resources are very limited. As you saw here, for certain problems they can perform just as well as a pretrained model.\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}