{"cells":[{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"cell_type":"code","source":"# This Python 3 environment comes with many helpful analytics libraries installed\n# It is defined by the kaggle/python Docker image: https://github.com/kaggle/docker-python\n# For example, here's several helpful packages to load\n\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n\n# Input data files are available in the read-only \"../input/\" directory\n# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory\n\nimport os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        print(os.path.join(dirname, filename))\n\n# You can write up to 5GB to the current directory (/kaggle/working/) that gets preserved as output when you create a version using \"Save & Run All\" \n# You can also write temporary files to /kaggle/temp/, but they won't be saved outside of the current session","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"from keras.preprocessing.image import ImageDataGenerator \nimport matplotlib\nfrom matplotlib import pyplot as plt\nimport matplotlib.image as mpimg\nimport tensorflow as tf","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train = \"/kaggle/input/dogs-vs-cats/train.zip\"\nworking = \"/kaggle/working\"","execution_count":null,"outputs":[]},{"metadata":{"_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","trusted":true},"cell_type":"code","source":"import zipfile\nwith zipfile.ZipFile(train, 'r') as zip_ref:\n    zip_ref.extractall(working)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_dir = os.path.join(working, \"train\")\nvalidation_dir = os.path.join(working, \"validation\")\n\nos.makedirs(os.path.join(train_dir, \"dogs\"))\nos.makedirs(os.path.join(train_dir, \"cats\"))\n\nos.makedirs(os.path.join(validation_dir, \"dogs\"))\nos.makedirs(os.path.join(validation_dir, \"cats\"))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"import shutil\nimport random\n\nsplit = 0.7\n\nfor file in os.listdir(train_dir):\n    if random.random() > split:\n        dest_dir = validation_dir\n    else:\n        dest_dir = train_dir\n    if file.endswith(\"jpg\"):  \n        if file.startswith(\"dog\"):\n            shutil.move(os.path.join(train_dir, file), os.path.join(dest_dir, \"dogs\"))\n        else:\n            shutil.move(os.path.join(train_dir, file), os.path.join(dest_dir, \"cats\"))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"classes = [\"cats\", \"dogs\"]\n\nfor cl in classes:\n    print(\"%s: %d\" % (cl, len(os.listdir(os.path.join(train_dir, cl)))))\n\nfor cl in classes:\n    print(\"%s: %d\" % (cl, len(os.listdir(os.path.join(validation_dir, cl)))))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"cats_dir = os.path.join(train_dir, \"cats\")\ndogs_dir = os.path.join(train_dir, \"dogs\")","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"%matplotlib inline","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"cats_dir = os.path.join(train_dir, \"cats\")\ncat_file_basename = os.listdir(cats_dir)[10]\ncat_file = os.path.join(cats_dir, cat_file_basename)\n\nplt.imshow(mpimg.imread(cat_file))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def check_empty_images(d):\n    for basename in os.listdir(d):\n        file = os.path.join(d, basename)\n        if os.path.getsize(file) == 0:\n            print(\"found bad\")\ncheck_empty_images(cats_dir)\ncheck_empty_images(dogs_dir)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"target_dimension = 200\nbatch_size = 10\ntrain_size = 0\nfor c in classes:\n    train_size += len(os.listdir(os.path.join(train_dir, c)))\nprint(\"train size: %d\" % train_size)\n\nvalidation_size = 0\nfor c in classes:\n    validation_size += len(os.listdir(os.path.join(validation_dir, c)))\nprint(\"validation size: %d\" % validation_size)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"import time\n\ndef train_model(model, epochs, train=None):\n    es = tf.keras.callbacks.EarlyStopping(monitor='val_accuracy', mode='max', verbose=1, patience=5, min_delta=0.001)\n    if train is None:\n        train = ImageDataGenerator(rescale=1./255)\n    train = train.flow_from_directory(\n        train_dir,\n        target_size=(target_dimension, target_dimension),\n        class_mode=\"binary\",\n        batch_size=batch_size,\n    )\n\n    validation = ImageDataGenerator(rescale=1./255)\n    validation = validation.flow_from_directory(\n        validation_dir,\n        target_size=(target_dimension, target_dimension),\n        class_mode=\"binary\",\n        batch_size=batch_size,\n    )\n\n    model.compile(optimizer=\"adam\", metrics=[\"accuracy\"], loss=\"binary_crossentropy\")\n    start = time.time()\n    history = model.fit_generator(\n        train,\n        steps_per_epoch=train_size / batch_size,\n        epochs=epochs,\n        validation_data=validation,\n        validation_steps=validation_size / batch_size,\n        #callbacks=[es],\n    )\n    print(\"training took: %f secs\" % (time.time() - start))\n    print(\"accuracy %f, validation accuracy: %f\" % (history.history['accuracy'][-1], history.history['val_accuracy'][-1]))\n\n    plt.title(\"Accuracy\")\n    plt.plot(range(0, epochs), history.history['accuracy'], label=\"training accuracy\")\n    plt.plot(range(0, epochs), history.history['val_accuracy'], label=\"validation accuracy\")\n    plt.legend()\n    plt.show()\n\n    plt.title(\"Loss\")\n    plt.plot(range(0, epochs), history.history['loss'], label=\"training loss\")\n    plt.plot(range(0, epochs), history.history['val_loss'], label=\"validation loss\")\n    plt.legend()\n    plt.show()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"model = tf.keras.models.Sequential([\n    tf.keras.layers.Conv2D(filters=32, kernel_size=(3, 3), input_shape=(target_dimension, target_dimension, 3), activation=\"relu\"),\n    tf.keras.layers.MaxPool2D(pool_size=(3, 3)),\n    tf.keras.layers.Dropout(0.5),\n    tf.keras.layers.Conv2D(filters=64, kernel_size=(3, 3), activation=\"relu\"),\n    tf.keras.layers.MaxPool2D(pool_size=(3, 3)),\n    tf.keras.layers.Dropout(0.5),\n    tf.keras.layers.Conv2D(filters=128, kernel_size=(3, 3), activation=\"relu\"),\n    tf.keras.layers.MaxPool2D(pool_size=(3, 3)),\n    tf.keras.layers.Dropout(0.5),\n    tf.keras.layers.Flatten(),\n    tf.keras.layers.Dense(100, activation=\"relu\"),\n    tf.keras.layers.Dropout(0.5),\n    tf.keras.layers.Dense(1, activation=\"sigmoid\")\n])\ntrain = ImageDataGenerator(\n    rescale=1./255,\n    horizontal_flip=True,\n)\ntrain_model(model, epochs=80, train=train)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]}],"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat":4,"nbformat_minor":4}