{ "cells": [ { "cell_type": "markdown", "metadata": { "colab_type": "text", "id": "x4HI2mpwlrcn" }, "source": [ "##### Copyright 2019 The TensorFlow Authors." ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "cellView": "form", "colab": {}, "colab_type": "code", "id": "679Lmwt3l1Bk" }, "outputs": [], "source": [ "#@title Licensed under the Apache License, Version 2.0 (the \"License\");\n", "# you may not use this file except in compliance with the License.\n", "# You may obtain a copy of the License at\n", "#\n", "# https://www.apache.org/licenses/LICENSE-2.0\n", "#\n", "# Unless required by applicable law or agreed to in writing, software\n", "# distributed under the License is distributed on an \"AS IS\" BASIS,\n", "# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n", "# See the License for the specific language governing permissions and\n", "# limitations under the License." ] }, { "cell_type": "markdown", "metadata": { "colab_type": "text", "id": "DSPCom-KmApV" }, "source": [ "# Convolutional Neural Network (CNN)" ] }, { "cell_type": "markdown", "metadata": { "colab_type": "text", "id": "klAltGp8ycek" }, "source": [ "\n", " \n", " \n", " \n", " \n", "
\n", " \n", " \n", " View on TensorFlow.org\n", " \n", " \n", " \n", " Run in Google Colab\n", " \n", " \n", " \n", " View source on GitHub\n", " \n", " Download notebook\n", "
" ] }, { "cell_type": "markdown", "metadata": { "colab_type": "text", "id": "qLGkt5qiyz4E" }, "source": [ "This tutorial demonstrates training a simple [Convolutional Neural Network](https://developers.google.com/machine-learning/glossary/#convolutional_neural_network) (CNN) to classify [CIFAR images](https://www.cs.toronto.edu/~kriz/cifar.html). Because this tutorial uses the [Keras Sequential API](https://www.tensorflow.org/guide/keras/overview), creating and training our model will take just a few lines of code.\n" ] }, { "cell_type": "markdown", "metadata": { "colab_type": "text", "id": "m7KBpffWzlxH" }, "source": [ "### Import TensorFlow" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "colab": {}, "colab_type": "code", "id": "iAve6DCL4JH4" }, "outputs": [], "source": [ "import tensorflow as tf\n", "\n", "from tensorflow.keras import datasets, layers, models\n", "import matplotlib.pyplot as plt" ] }, { "cell_type": "markdown", "metadata": { "colab_type": "text", "id": "jRFxccghyMVo" }, "source": [ "### Download and prepare the CIFAR10 dataset\n", "\n", "\n", "The CIFAR10 dataset contains 60,000 color images in 10 classes, with 6,000 images in each class. The dataset is divided into 50,000 training images and 10,000 testing images. The classes are mutually exclusive and there is no overlap between them." ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "colab": {}, "colab_type": "code", "id": "JWoEqyMuXFF4" }, "outputs": [], "source": [ "(train_images, train_labels), (test_images, test_labels) = datasets.cifar10.load_data()\n", "\n", "# Normalize pixel values to be between 0 and 1\n", "train_images, test_images = train_images / 255.0, test_images / 255.0" ] }, { "cell_type": "markdown", "metadata": { "colab_type": "text", "id": "7wArwCTJJlUa" }, "source": [ "### Verify the data\n", "\n", "To verify that the dataset looks correct, let's plot the first 25 images from the training set and display the class name below each image.\n" ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "colab": {}, "colab_type": "code", "id": "K3PAELE2eSU9" }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "class_names = ['airplane', 'automobile', 'bird', 'cat', 'deer',\n", " 'dog', 'frog', 'horse', 'ship', 'truck']\n", "\n", "plt.figure(figsize=(10,10))\n", "for i in range(25):\n", " plt.subplot(5,5,i+1)\n", " plt.xticks([])\n", " plt.yticks([])\n", " plt.grid(False)\n", " plt.imshow(train_images[i], cmap=plt.cm.binary)\n", " # The CIFAR labels happen to be arrays, \n", " # which is why you need the extra index\n", " plt.xlabel(class_names[train_labels[i][0]])\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": { "colab_type": "text", "id": "Oewp-wYg31t9" }, "source": [ "### Create the convolutional base" ] }, { "cell_type": "markdown", "metadata": { "colab_type": "text", "id": "3hQvqXpNyN3x" }, "source": [ "The 6 lines of code below define the convolutional base using a common pattern: a stack of [Conv2D](https://www.tensorflow.org/api_docs/python/tf/keras/layers/Conv2D) and [MaxPooling2D](https://www.tensorflow.org/api_docs/python/tf/keras/layers/MaxPool2D) layers.\n", "\n", "As input, a CNN takes tensors of shape (image_height, image_width, color_channels), ignoring the batch size. If you are new to these dimensions, color_channels refers to (R,G,B). In this example, you will configure our CNN to process inputs of shape (32, 32, 3), which is the format of CIFAR images. You can do this by passing the argument `input_shape` to our first layer.\n" ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "colab": {}, "colab_type": "code", "id": "L9YmGQBQPrdn" }, "outputs": [], "source": [ "model = models.Sequential()\n", "model.add(layers.Conv2D(32, (3, 3), activation='relu', input_shape=(32, 32, 3)))\n", "model.add(layers.MaxPooling2D((2, 2)))\n", "model.add(layers.Conv2D(64, (3, 3), activation='relu'))\n", "model.add(layers.MaxPooling2D((2, 2)))\n", "model.add(layers.Conv2D(64, (3, 3), activation='relu'))" ] }, { "cell_type": "markdown", "metadata": { "colab_type": "text", "id": "lvDVFkg-2DPm" }, "source": [ "Let's display the architecture of our model so far." ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "colab": {}, "colab_type": "code", "id": "8-C4XBg4UTJy" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Model: \"sequential\"\n", "_________________________________________________________________\n", "Layer (type) Output Shape Param # \n", "=================================================================\n", "conv2d (Conv2D) (None, 30, 30, 32) 896 \n", "_________________________________________________________________\n", "max_pooling2d (MaxPooling2D) (None, 15, 15, 32) 0 \n", "_________________________________________________________________\n", "conv2d_1 (Conv2D) (None, 13, 13, 64) 18496 \n", "_________________________________________________________________\n", "max_pooling2d_1 (MaxPooling2 (None, 6, 6, 64) 0 \n", "_________________________________________________________________\n", "conv2d_2 (Conv2D) (None, 4, 4, 64) 36928 \n", "=================================================================\n", "Total params: 56,320\n", "Trainable params: 56,320\n", "Non-trainable params: 0\n", "_________________________________________________________________\n" ] } ], "source": [ "model.summary()" ] }, { "cell_type": "markdown", "metadata": { "colab_type": "text", "id": "_j-AXYeZ2GO5" }, "source": [ "Above, you can see that the output of every Conv2D and MaxPooling2D layer is a 3D tensor of shape (height, width, channels). The width and height dimensions tend to shrink as you go deeper in the network. The number of output channels for each Conv2D layer is controlled by the first argument (e.g., 32 or 64). Typically, as the width and height shrink, you can afford (computationally) to add more output channels in each Conv2D layer." ] }, { "cell_type": "markdown", "metadata": { "colab_type": "text", "id": "_v8sVOtG37bT" }, "source": [ "### Add Dense layers on top\n", "To complete our model, you will feed the last output tensor from the convolutional base (of shape (4, 4, 64)) into one or more Dense layers to perform classification. Dense layers take vectors as input (which are 1D), while the current output is a 3D tensor. First, you will flatten (or unroll) the 3D output to 1D, then add one or more Dense layers on top. CIFAR has 10 output classes, so you use a final Dense layer with 10 outputs and a softmax activation." ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "colab": {}, "colab_type": "code", "id": "mRs95d6LUVEi" }, "outputs": [], "source": [ "model.add(layers.Flatten())\n", "model.add(layers.Dense(64, activation='relu'))\n", "model.add(layers.Dense(10))" ] }, { "cell_type": "markdown", "metadata": { "colab_type": "text", "id": "ipGiQMcR4Gtq" }, "source": [ "Here's the complete architecture of our model." ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "colab": {}, "colab_type": "code", "id": "8Yu_m-TZUWGX" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Model: \"sequential\"\n", "_________________________________________________________________\n", "Layer (type) Output Shape Param # \n", "=================================================================\n", "conv2d (Conv2D) (None, 30, 30, 32) 896 \n", "_________________________________________________________________\n", "max_pooling2d (MaxPooling2D) (None, 15, 15, 32) 0 \n", "_________________________________________________________________\n", "conv2d_1 (Conv2D) (None, 13, 13, 64) 18496 \n", "_________________________________________________________________\n", "max_pooling2d_1 (MaxPooling2 (None, 6, 6, 64) 0 \n", "_________________________________________________________________\n", "conv2d_2 (Conv2D) (None, 4, 4, 64) 36928 \n", "_________________________________________________________________\n", "flatten (Flatten) (None, 1024) 0 \n", "_________________________________________________________________\n", "dense (Dense) (None, 64) 65600 \n", "_________________________________________________________________\n", "dense_1 (Dense) (None, 10) 650 \n", "=================================================================\n", "Total params: 122,570\n", "Trainable params: 122,570\n", "Non-trainable params: 0\n", "_________________________________________________________________\n" ] } ], "source": [ "model.summary()" ] }, { "cell_type": "markdown", "metadata": { "colab_type": "text", "id": "xNKXi-Gy3RO-" }, "source": [ "As you can see, our (4, 4, 64) outputs were flattened into vectors of shape (1024) before going through two Dense layers." ] }, { "cell_type": "markdown", "metadata": { "colab_type": "text", "id": "P3odqfHP4M67" }, "source": [ "### Compile and train the model" ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "colab": {}, "colab_type": "code", "id": "MdDzI75PUXrG" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1/10\n", "1563/1563 [==============================] - 33s 21ms/step - loss: 1.5091 - accuracy: 0.4496 - val_loss: 1.2413 - val_accuracy: 0.5544\n", "Epoch 2/10\n", "1563/1563 [==============================] - 36s 23ms/step - loss: 1.1461 - accuracy: 0.5938 - val_loss: 1.0786 - val_accuracy: 0.6183\n", "Epoch 3/10\n", "1563/1563 [==============================] - 33s 21ms/step - loss: 0.9955 - accuracy: 0.6507 - val_loss: 0.9655 - val_accuracy: 0.6618\n", "Epoch 4/10\n", "1563/1563 [==============================] - 34s 22ms/step - loss: 0.8931 - accuracy: 0.6870 - val_loss: 0.9757 - val_accuracy: 0.6678\n", "Epoch 5/10\n", "1563/1563 [==============================] - 34s 22ms/step - loss: 0.8159 - accuracy: 0.7125 - val_loss: 0.8976 - val_accuracy: 0.6881\n", "Epoch 6/10\n", "1563/1563 [==============================] - 36s 23ms/step - loss: 0.7606 - accuracy: 0.7345 - val_loss: 0.8738 - val_accuracy: 0.6974\n", "Epoch 7/10\n", "1563/1563 [==============================] - 35s 23ms/step - loss: 0.7111 - accuracy: 0.7498 - val_loss: 0.8657 - val_accuracy: 0.7099\n", "Epoch 8/10\n", "1563/1563 [==============================] - 33s 21ms/step - loss: 0.6617 - accuracy: 0.7676 - val_loss: 0.8830 - val_accuracy: 0.7053\n", "Epoch 9/10\n", "1563/1563 [==============================] - 36s 23ms/step - loss: 0.6232 - accuracy: 0.7815 - val_loss: 0.8630 - val_accuracy: 0.7135\n", "Epoch 10/10\n", "1563/1563 [==============================] - 39s 25ms/step - loss: 0.5825 - accuracy: 0.7953 - val_loss: 0.9164 - val_accuracy: 0.6942\n" ] } ], "source": [ "model.compile(optimizer='adam',\n", " loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),\n", " metrics=['accuracy'])\n", "\n", "history = model.fit(train_images, train_labels, epochs=10, \n", " validation_data=(test_images, test_labels))" ] }, { "cell_type": "markdown", "metadata": { "colab_type": "text", "id": "jKgyC5K_4O0d" }, "source": [ "### Evaluate the model" ] }, { "cell_type": "raw", "metadata": { "colab": {}, "colab_type": "code", "id": "gtyDF0MKUcM7" }, "source": [ "plt.plot(history.history['accuracy'], label='accuracy')\n", "plt.plot(history.history['val_accuracy'], label = 'val_accuracy')\n", "plt.xlabel('Epoch')\n", "plt.ylabel('Accuracy')\n", "plt.ylim([0.5, 1])\n", "plt.legend(loc='lower right')\n", "\n", "test_loss, test_acc = model.evaluate(test_images, test_labels, verbose=2)" ] }, { "cell_type": "code", "execution_count": 10, "metadata": { "colab": {}, "colab_type": "code", "id": "0LvwaKhtUdOo" }, "outputs": [ { "ename": "NameError", "evalue": "name 'test_acc' is not defined", "output_type": "error", "traceback": [ "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", "\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)", "\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtest_acc\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m", "\u001b[0;31mNameError\u001b[0m: name 'test_acc' is not defined" ] } ], "source": [ "print(test_acc)" ] }, { "cell_type": "markdown", "metadata": { "colab_type": "text", "id": "8cfJ8AR03gT5" }, "source": [ "Our simple CNN has achieved a test accuracy of over 70%. Not bad for a few lines of code! For another CNN style, see an example using the Keras subclassing API and a `tf.GradientTape` [here](https://www.tensorflow.org/tutorials/quickstart/advanced)." ] } ], "metadata": { "accelerator": "GPU", "colab": { "collapsed_sections": [], "name": "cnn.ipynb", "private_outputs": true, "provenance": [], "toc_visible": true }, "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.7.4" } }, "nbformat": 4, "nbformat_minor": 1 }