{"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":{"_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","trusted":true},"cell_type":"code","source":"import tensorflow as tf\nfrom kaggle_datasets import KaggleDatasets\nimport math, re, os, random\nimport numpy as np\nfrom keras.preprocessing.image import ImageDataGenerator\nfrom sklearn.model_selection import StratifiedKFold\nfrom tensorflow.keras.callbacks import ReduceLROnPlateau\nfrom tensorflow.keras.optimizers import Adam\n\nprint(\"Tensorflow version \" + tf.__version__)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"try:\n    tpu = tf.distribute.cluster_resolver.TPUClusterResolver()  # TPU detection. No parameters necessary if TPU_NAME environment variable is set. On Kaggle this is always the case.\n    print('Running on TPU ', tpu.master())\nexcept ValueError:\n    tpu = None\n\nif tpu:\n    tf.config.experimental_connect_to_cluster(tpu)\n    tf.tpu.experimental.initialize_tpu_system(tpu)\n    strategy = tf.distribute.experimental.TPUStrategy(tpu)\nelse:\n    strategy = tf.distribute.get_strategy() # default distribution strategy in Tensorflow. Works on CPU and single GPU.\n\nprint(\"REPLICAS: \", strategy.num_replicas_in_sync)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"GCS_DS_PATH = KaggleDatasets().get_gcs_path() # you can list the bucket with \"!gsutil ls $GCS_DS_PATH\"","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"IMAGE_SIZE = [224, 224] # at this size, a GPU will run out of memory. Use the TPU\nEPOCHS = 70\nBATCH_SIZE = 16 * strategy.num_replicas_in_sync\nSEED = 42\nNUM_TRAINING_IMAGES = 12753\nNUM_TEST_IMAGES = 7382\nSTEPS_PER_EPOCH = NUM_TRAINING_IMAGES // BATCH_SIZE","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def random_blockout(img, sl = 0.1, sh = 0.2, rl = 0.4):\n    p = random.random()\n    if p >= 0.25:\n        w, h, c = IMAGE_SIZE[0], IMAGE_SIZE[1], 3\n        origin_area = tf.cast(h * w, tf.float32)\n\n        e_size_l = tf.cast(tf.round(tf.sqrt(origin_area * sl * rl)), tf.int32)\n        e_size_h = tf.cast(tf.round(tf.sqrt(origin_area * sh / rl)), tf.int32)\n\n        e_height_h = tf.minimum(e_size_h, h)\n        e_width_h = tf.minimum(e_size_h, w)\n\n        erase_height = tf.random.uniform(shape = [], minval = e_size_l, maxval = e_height_h, dtype = tf.int32)\n        erase_width = tf.random.uniform(shape = [], minval = e_size_l, maxval = e_width_h, dtype = tf.int32)\n\n        erase_area = tf.zeros(shape = [erase_height, erase_width, c])\n        erase_area = tf.cast(erase_area, tf.uint8)\n\n        pad_h = h - erase_height\n        pad_top = tf.random.uniform(shape = [], minval = 0, maxval = pad_h, dtype = tf.int32)\n        pad_bottom = pad_h - pad_top\n\n        pad_w = w - erase_width\n        pad_left = tf.random.uniform(shape = [], minval = 0, maxval = pad_w, dtype = tf.int32)\n        pad_right = pad_w - pad_left\n\n        erase_mask = tf.pad([erase_area], [[0,0],[pad_top, pad_bottom], [pad_left, pad_right], [0,0]], constant_values=1)\n        erase_mask = tf.squeeze(erase_mask, axis = 0)\n        erased_img = tf.multiply(tf.cast(img,tf.float32), tf.cast(erase_mask, tf.float32))\n\n        return tf.cast(erased_img, img.dtype)\n    else:\n        return tf.cast(img, img.dtype)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def decode_image(image_data):\n    image = tf.image.decode_jpeg(image_data, channels=3)\n    image = tf.cast(image, tf.float32) / 255.0  # convert image to floats in [0, 1] range\n    image = tf.reshape(image, [*IMAGE_SIZE, 3]) # explicit size needed for TPU\n    return image\n\ndef read_labeled_tfrecord(example):\n    LABELED_TFREC_FORMAT = {\n        \"image\": tf.io.FixedLenFeature([], tf.string), # tf.string means bytestring\n        \"class\": tf.io.FixedLenFeature([], tf.int64),  # shape [] means single element\n    }\n    example = tf.io.parse_single_example(example, LABELED_TFREC_FORMAT)\n    image = decode_image(example['image'])\n    label = tf.cast(example['class'], tf.int32)\n    return image, label # returns a dataset of (image, label) pairs\n\ndef read_unlabeled_tfrecord(example):\n    UNLABELED_TFREC_FORMAT = {\n        \"image\": tf.io.FixedLenFeature([], tf.string), # tf.string means bytestring\n        \"id\": tf.io.FixedLenFeature([], tf.string),  # shape [] means single element\n        # class is missing, this competitions's challenge is to predict flower classes for the test dataset\n    }\n    example = tf.io.parse_single_example(example, UNLABELED_TFREC_FORMAT)\n    image = decode_image(example['image'])\n    idnum = example['id']\n    return image, idnum # returns a dataset of image(s)\n\ndef load_dataset(filenames, labeled=True, ordered=False):\n    # Read from TFRecords. For optimal performance, reading from multiple files at once and\n    # disregarding data order. Order does not matter since we will be shuffling the data anyway.\n\n    ignore_order = tf.data.Options()\n    if not ordered:\n        ignore_order.experimental_deterministic = False # disable order, increase speed\n\n    dataset = tf.data.TFRecordDataset(filenames) # automatically interleaves reads from multiple files\n    dataset = dataset.with_options(ignore_order) # uses data as soon as it streams in, rather than in its original order\n    dataset = dataset.map(read_labeled_tfrecord if labeled else read_unlabeled_tfrecord)\n    # returns a dataset of (image, label) pairs if labeled=True or (image, id) pairs if labeled=False\n    return dataset\n\ndef data_augment(image, label):\n    # data augmentation. Thanks to the dataset.prefetch(AUTO) statement in the next function (below),\n    # this happens essentially for free on TPU. Data pipeline code is executed on the \"CPU\" part\n    # of the TPU while the TPU itself is computing gradients.\n\n    image = tf.image.random_flip_left_right(image, seed=SEED)\n    image = random_blockout(image)\n    return image, label\n\ndef get_training_dataset():\n    dataset = load_dataset(tf.io.gfile.glob(GCS_DS_PATH + '/tfrecords-jpeg-224x224/train/*.tfrec'), labeled=True)\n    dataset = dataset.map(data_augment)\n    dataset = dataset.repeat() # the training dataset must repeat for several epochs\n    dataset = dataset.shuffle(2048)\n    dataset = dataset.batch(BATCH_SIZE)\n    return dataset\n\ndef get_validation_dataset():\n    dataset = load_dataset(tf.io.gfile.glob(GCS_DS_PATH + '/tfrecords-jpeg-224x224/val/*.tfrec'), labeled=True, ordered=False)\n    dataset = dataset.batch(BATCH_SIZE)\n    dataset = dataset.cache()\n    return dataset\n\ndef get_test_dataset(ordered=False):\n    dataset = load_dataset(tf.io.gfile.glob(GCS_DS_PATH + '/tfrecords-jpeg-224x224/test/*.tfrec'), labeled=False, ordered=ordered)\n    dataset = dataset.batch(BATCH_SIZE)\n    return dataset","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# !pip install -q efficientnet\n# import efficientnet.tfkeras as efn","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"learning_rate_reduction = ReduceLROnPlateau(monitor = 'val_sparse_categorical_accuracy', patience = 3, verbose = 1, \n                                           factor = 0.2, min_lr = 0.00001)\n\noptimizer = Adam(lr = .0001, beta_1 = .9, beta_2 = .999, epsilon = None, decay = .0, amsgrad = False)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"training_dataset = get_training_dataset()\nvalidation_dataset = get_validation_dataset()\n\nwith strategy.scope():\n    pretrained_model = tf.keras.applications.ResNet50V2(\n    weights='imagenet',\n    include_top=False ,\n    input_shape=[*IMAGE_SIZE, 3]\n    )\n    pretrained_model.trainable = False\n    \n    model = tf.keras.Sequential([\n        # To a base pretrained on ImageNet to extract features from images...\n        pretrained_model,\n        # ... attach a new head to act as a classifier.\n        tf.keras.layers.GlobalAveragePooling2D(),\n        tf.keras.layers.Dense(104, activation='softmax')\n    ])\n    model.compile(\n        optimizer='adam',\n        loss = 'sparse_categorical_crossentropy',\n        metrics=['sparse_categorical_accuracy'],\n    )\n\n# model.summary()\n\n# with strategy.scope():    \n#     pretrained_model =  efn.EfficientNetB0(weights = 'imagenet', include_top = False , input_shape = [*IMAGE_SIZE, 3])\n#     pretrained_model.trainable = False # tramsfer learning\n    \n#     model = tf.keras.Sequential([\n#         pretrained_model,\n#         tf.keras.layers.GlobalAveragePooling2D(),\n#         tf.keras.layers.Dense(104, activation='softmax')\n#     ])\n        \n# model.compile(\n#     optimizer = optimizer,\n#     loss = 'sparse_categorical_crossentropy',\n#     metrics=['sparse_categorical_accuracy']\n# )\n\nhistorical = model.fit(training_dataset, \n          steps_per_epoch=STEPS_PER_EPOCH, \n          epochs = EPOCHS,\n          callbacks = [learning_rate_reduction],\n          validation_data=validation_dataset)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# historical.history['loss'].plot()\ndiff = pd.DataFrame({'loss': [historical.history['loss']]})","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"diff","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"test_ds = get_test_dataset(ordered = True) # since we are splitting the dataset and iterating separately on images and ids, order matters.\n\nprint('Computing predictions...')\ntest_images_ds = test_ds.map(lambda image, idnum: image)\nprobabilities = model.predict(test_images_ds)\npredictions = np.argmax(probabilities, axis=-1)\nprint(predictions)\n\nprint('Generating submission.csv file...')\ntest_ids_ds = test_ds.map(lambda image, idnum: idnum).unbatch()\ntest_ids = next(iter(test_ids_ds.batch(NUM_TEST_IMAGES))).numpy().astype('U') # all in one batch\nnp.savetxt('submission.csv', np.rec.fromarrays([test_ids, predictions]), fmt = ['%s', '%d'], delimiter = ',', header = 'id,label', comments = '')","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}