Skip to content
Snippets Groups Projects
Jupyter Notebook Block 5 - Object Detection and Segmentation.ipynb 692 KiB
Newer Older
{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "dWyPGNkCGhIX"
   },
   "source": [
    "# Part I : Create Your Own Dataset and Train it with ConvNets\n",
    "\n",
    "In this part of the notebook, you will set up your own dataset for image classification. Please specify \n",
    "under `queries` the image categories you are interested in. Under `limit` specify the number of images \n",
    "you want to download for each image category. \n",
    "\n",
    "You do not need to understand the class `simple_image_download`, just execute the cell after you have specified \n",
    "the download folder.\n"
   ]
  },
  {
   "cell_type": "code",
Simon van Hemert's avatar
Simon van Hemert committed
   "execution_count": 1,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "8rckz3ZuGhIc",
    "outputId": "6f615f06-759a-4eea-839e-658155df8d36"
   },
Simon van Hemert's avatar
Simon van Hemert committed
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 2 image links\n",
      "Saved 2 images\n",
      "Found 2 image links\n",
      "Saved 2 images\n"
Simon van Hemert's avatar
Simon van Hemert committed
     ]
   "source": [
Simon van Hemert's avatar
Simon van Hemert committed
    "from selenium import webdriver\n",
    "from selenium.webdriver.firefox.options import Options\n",
    "from Image_crawling import Image_crawling\n",
    "\n",
    "# Specifiy the queries\n",
    "queries = [\"brad pitt\",\"johnny depp\", \"leonardo dicaprio\", \"robert de niro\", \"angelina jolie\", \"sandra bullock\", \"catherine deneuve\", \"marion cotillard\"]\n",
    "#queries = [\"Bart Simpson\",\"Homer Simpson\"]\n",
    "limit = 2\n",
Simon van Hemert's avatar
Simon van Hemert committed
    "download_folder = \"./brandnew_images/train/\"\n",
Simon van Hemert's avatar
Simon van Hemert committed
    "waittime = 0.1  # Time to wait between actions, depends on the number of pictures you want to crawl. More pictures means you need to wait longer for them to load. \n",
Simon van Hemert's avatar
Simon van Hemert committed
    "# Set options\n",
    "options = webdriver.FirefoxOptions()\n",
    "options.add_argument('--headless')\n",
Simon van Hemert's avatar
Simon van Hemert committed
    "# Create Driver\n",
    "driver = webdriver.Firefox(options=options, executable_path=\"/usr/bin/geckodriver\")\n",
Simon van Hemert's avatar
Simon van Hemert committed
    "# create instance of crawler\n",
    "image_crawling = Image_crawling(driver, waittime=waittime)\n",
Simon van Hemert's avatar
Simon van Hemert committed
    "# Find urls and download images\n",
Simon van Hemert's avatar
Simon van Hemert committed
    "for query in queries:\n",
    "    # Craws image urls:\n",
    "    image_urls = image_crawling.fetch_image_urls(query, limit)\n",
    "      \n",
    "    # download images\n",
    "    image_crawling.download_image(download_folder + query)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "CRHl9UX6GhIs"
   },
   "source": [
    "Please check carefully the downloaded images, there may be a lot of garbage! You definitely need to \n",
    "clean the data.\n",
    "\n",
    "In the following, you will apply data augmentation to your data set."
   ]
  },
  {
   "cell_type": "code",
Mirko Birbaumer's avatar
Mirko Birbaumer committed
   "execution_count": 2,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "3SX21FtcGhIu"
   },
   "outputs": [],
   "source": [
    "# General imports\n",
    "import tensorflow as tf\n",
    "tf.compat.v1.enable_eager_execution(\n",
    "    config=None, device_policy=None, execution_mode=None\n",
    ")\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Shortcuts to keras if (however from tensorflow)\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Conv2D, MaxPooling2D\n",
    "from tensorflow.keras.layers import Activation, Dropout, Flatten, Dense\n",
    "from tensorflow.keras.callbacks import TensorBoard \n",
    "\n",
    "# Shortcut for displaying images\n",
    "def plot_img(img):\n",
    "    plt.imshow(img, cmap='gray')\n",
    "    plt.axis(\"off\")\n",
    "    plt.show()\n",
    "    \n",
    "# The target image size can be fixed here (quadratic)\n",
    "# the ImageDataGenerator() automatically scales the images accordingly (aspect ratio is changed)\n",
    "image_size = 150"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "rN_Mp1rmGhI1",
    "outputId": "6417b1f9-e7d4-4d56-a213-191f9d17524a"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 480 images belonging to 8 classes.\n"
Mirko Birbaumer's avatar
Mirko Birbaumer committed
     "data": {
      "image/png": "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\n",
Mirko Birbaumer's avatar
Mirko Birbaumer committed
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "array([0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "       0., 0., 0., 0., 0., 0., 0., 0.], dtype=float32)"
     "execution_count": 4,
Mirko Birbaumer's avatar
Mirko Birbaumer committed
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# These are the class names; this defines the ordering of the classes\n",
Simon van Hemert's avatar
Simon van Hemert committed
    "class_names = [\"brad pitt\", \"johnny depp\", \"leonardo dicaprio\", \"robert de niro\",\n",
    "           \"angelina jolie\", \"sandra bullock\", \"catherine deneuve\", \"marion cotillard\"]\n",
    "\n",
    "\n",
    "# Class ImageDataGenerator() returns an iterator holding one batch of images\n",
    "# the constructor takes arguments defining the different image transformations\n",
    "# for augmentation purposes (rotation, x-/y-shift, intensity scaling - here 1./255 \n",
    "# to scale range to [0, 1], shear, zoom, flip, ... )\n",
    "train_datagen = ImageDataGenerator(\n",
    "        rotation_range=10,\n",
    "        width_shift_range=0.2,\n",
    "        height_shift_range=0.2,\n",
    "        rescale=1./255,\n",
    "        shear_range=0.2,\n",
    "        zoom_range=0.2,\n",
    "        horizontal_flip=True,\n",
    "        fill_mode='nearest')\n",
    "\n",
    "\n",
    "dir_iter = train_datagen.flow_from_directory('./train/', \n",
    "                                         target_size=(image_size, image_size),\n",
    "                                         classes=class_names,\n",
    "                                         batch_size=25, class_mode='sparse', shuffle=False)\n",
    "\n",
Simon van Hemert's avatar
Simon van Hemert committed
    "plot_img(dir_iter[0][0][1,...])\n",
    "dir_iter[0][1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "V2fYccc8GhJF"
   },
   "source": [
    "Before you continue, you need to split the downloaded images into a `train` folder and into a `validation` folder."
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "colab_type": "raw",
    "id": "VamXG4FoGhJH"
   },
   "source": [
    "./\n",
    "├── train\n",
    "│   ├── brad pitt\n",
    "│   └── johnny deep\n",
    "|   ├── leonardo di caprio\n",
    "|   └── ...\n",
    "│       \n",
    "└── validation\n",
    "    ├── brad pitt\n",
    "    ├── johnny deep\n",
    "    ├── leonardo di caprio\n",
    "    └── ..."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "9322su6vGhJJ"
   },
   "source": [
Simon van Hemert's avatar
Simon van Hemert committed
    "If you want to use the example of this jupyter notebook, you can use the images provided in the ./train and ./validation folders."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "xPqJWgeAGhJL"
   },
   "source": [
    "## Define a ConvNet Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "UuJV4JBKGhJO"
   },
   "outputs": [],
   "source": [
    "batch_size = 20\n",
    "num_train_images = 480\n",
    "num_valid_images = 80\n",
    "num_classes = 8\n",
    "\n",
    "model_scratch = Sequential()\n",
    "model_scratch.add(Conv2D(32, (3, 3), input_shape=(image_size, image_size, 3)))\n",
    "model_scratch.add(Activation('relu'))\n",
    "model_scratch.add(MaxPooling2D(pool_size=(2, 2)))\n",
    "\n",
    "model_scratch.add(Conv2D(32, (3, 3)))\n",
    "model_scratch.add(Activation('relu'))\n",
    "model_scratch.add(MaxPooling2D(pool_size=(2, 2)))\n",
    "\n",
    "model_scratch.add(Conv2D(64, (3, 3)))\n",
    "model_scratch.add(Activation('relu'))\n",
    "model_scratch.add(MaxPooling2D(pool_size=(2, 2)))\n",
    "\n",
    "# this converts our 3D feature maps to 1D feature vectors\n",
    "model_scratch.add(Flatten())  \n",
    "model_scratch.add(Dense(64))\n",
    "model_scratch.add(Activation('relu'))\n",
    "model_scratch.add(Dropout(0.5))\n",
    "model_scratch.add(Dense(num_classes))\n",
    "model_scratch.add(Activation('softmax'))\n",
    "\n",
    "model_scratch.compile(loss='categorical_crossentropy',\n",
    "              optimizer='adam',\n",
    "              metrics=['accuracy'])\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "JFdkIokMGhJT",
    "outputId": "63e7d032-4083-4fe0-d970-c10bf0c39a94"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 480 images belonging to 8 classes.\n",
      "Found 80 images belonging to 8 classes.\n"
     ]
    }
   ],
   "source": [
    "# This is the augmentation configuration we will use for training\n",
    "train_datagen = ImageDataGenerator(\n",
    "        rescale=1./255,\n",
    "        shear_range=0.2,\n",
    "        zoom_range=0.2,\n",
    "        horizontal_flip=True)\n",
    "\n",
    "# This is the augmentation configuration we will use for validation:\n",
    "# only rescaling\n",
    "validation_datagen = ImageDataGenerator(rescale=1./255)\n",
    "\n",
    "# This is a generator that will read pictures found in\n",
    "# subfolers of './train', and indefinitely generate\n",
    "# batches of augmented image data\n",
    "train_generator = train_datagen.flow_from_directory(\n",
    "        './train',  # this is the target directory\n",
    "        target_size=(image_size, image_size),  # all images will be resized to 150x150\n",
    "        classes=class_names,\n",
    "        batch_size=batch_size)  \n",
    "\n",
    "# This is a similar generator, for validation data\n",
    "validation_generator = validation_datagen.flow_from_directory(\n",
    "        './validation',\n",
    "        target_size = (image_size, image_size),\n",
    "        classes = class_names,\n",
    "        batch_size = batch_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "cytHiQUTGhJb"
   },
   "outputs": [],
   "source": [
    "name = 'cnn_face_1'\n",
    "tensorboard_callback = tf.keras.callbacks.TensorBoard(log_dir ='./tensorboard/' + name + '/', \n",
    "                                                      histogram_freq=1, \n",
    "                                                      write_graph=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "C7dCbyXPGhJg",
    "outputId": "98b4085e-ed6d-43e2-831f-aec32161583f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "10/24 [===========>..................] - ETA: 26s - loss: 2.1636 - accuracy: 0.1200"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.7/site-packages/PIL/Image.py:952: UserWarning: Palette images with Transparency expressed in bytes should be converted to RGBA images\n",
      "  \"Palette images with Transparency expressed in bytes should be \"\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "24/24 [==============================] - 45s 2s/step - loss: 2.1158 - accuracy: 0.0979 - val_loss: 2.0759 - val_accuracy: 0.1625\n",
      "Epoch 2/20\n",
      "24/24 [==============================] - 38s 2s/step - loss: 2.0772 - accuracy: 0.1667 - val_loss: 2.0598 - val_accuracy: 0.3875\n",
      "Epoch 3/20\n",
      "24/24 [==============================] - 38s 2s/step - loss: 2.0419 - accuracy: 0.2062 - val_loss: 1.9549 - val_accuracy: 0.3250\n",
      "Epoch 4/20\n",
      "24/24 [==============================] - 38s 2s/step - loss: 1.9731 - accuracy: 0.2542 - val_loss: 1.8280 - val_accuracy: 0.3625\n",
      "Epoch 5/20\n",
      "24/24 [==============================] - 36s 1s/step - loss: 1.8939 - accuracy: 0.2458 - val_loss: 1.7558 - val_accuracy: 0.4250\n",
      "Epoch 6/20\n",
      "24/24 [==============================] - 36s 1s/step - loss: 1.8929 - accuracy: 0.2625 - val_loss: 1.6907 - val_accuracy: 0.3750\n",
      "Epoch 7/20\n",
      "24/24 [==============================] - 36s 1s/step - loss: 1.7998 - accuracy: 0.3042 - val_loss: 1.6398 - val_accuracy: 0.4250\n",
      "Epoch 8/20\n",
      "24/24 [==============================] - 35s 1s/step - loss: 1.7445 - accuracy: 0.3458 - val_loss: 1.6191 - val_accuracy: 0.4125\n",
      "Epoch 9/20\n",
      "24/24 [==============================] - 32s 1s/step - loss: 1.6618 - accuracy: 0.3688 - val_loss: 1.5744 - val_accuracy: 0.4000\n",
      "Epoch 10/20\n",
      "24/24 [==============================] - 34s 1s/step - loss: 1.6509 - accuracy: 0.3562 - val_loss: 1.6113 - val_accuracy: 0.3750\n",
      "Epoch 11/20\n",
      "24/24 [==============================] - 34s 1s/step - loss: 1.5888 - accuracy: 0.3854 - val_loss: 1.4971 - val_accuracy: 0.4750\n",
      "Epoch 12/20\n",
      "24/24 [==============================] - 34s 1s/step - loss: 1.5139 - accuracy: 0.4333 - val_loss: 1.4896 - val_accuracy: 0.5125\n",
      "Epoch 13/20\n",
      "24/24 [==============================] - 36s 1s/step - loss: 1.4464 - accuracy: 0.4646 - val_loss: 1.4945 - val_accuracy: 0.5250\n",
      "Epoch 14/20\n",
      "24/24 [==============================] - 34s 1s/step - loss: 1.4098 - accuracy: 0.5063 - val_loss: 1.4352 - val_accuracy: 0.5000\n",
      "Epoch 15/20\n",
      "24/24 [==============================] - 36s 1s/step - loss: 1.3459 - accuracy: 0.4938 - val_loss: 1.4270 - val_accuracy: 0.5000\n",
      "Epoch 16/20\n",
      "24/24 [==============================] - 35s 1s/step - loss: 1.3225 - accuracy: 0.5271 - val_loss: 1.4735 - val_accuracy: 0.4375\n",
      "Epoch 17/20\n",
      "24/24 [==============================] - 35s 1s/step - loss: 1.1904 - accuracy: 0.5771 - val_loss: 1.7510 - val_accuracy: 0.4625\n",
      "Epoch 18/20\n",
      "24/24 [==============================] - 34s 1s/step - loss: 1.4096 - accuracy: 0.4958 - val_loss: 1.4725 - val_accuracy: 0.4875\n",
      "Epoch 19/20\n",
      "24/24 [==============================] - 34s 1s/step - loss: 1.1918 - accuracy: 0.5146 - val_loss: 1.5643 - val_accuracy: 0.4375\n",
      "Epoch 20/20\n",
      "24/24 [==============================] - 35s 1s/step - loss: 1.1331 - accuracy: 0.5938 - val_loss: 1.5486 - val_accuracy: 0.4375\n"
     ]
    }
   ],
   "source": [
    "history = model_scratch.fit(\n",
Simon van Hemert's avatar
Simon van Hemert committed
    "    train_generator,\n",
    "    steps_per_epoch = num_train_images // batch_size,\n",
    "    epochs = 20,\n",
    "    validation_data = validation_generator,\n",
    "    validation_steps = num_valid_images // batch_size,\n",
    "    callbacks = [tensorboard_callback])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "wt_ONw5PGhJm",
    "outputId": "e75d8a73-da49-4dbe-ffcf-7cb316be39a2"
   },
   "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"
    },
    {
     "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.history['accuracy'])\n",
    "plt.plot(histor
Loading
Loading full blame...