{"cells":[{"metadata":{},"cell_type":"markdown","source":"## Recipe for accuracy 96+ \n0. Use the [getting started notebook](https://www.kaggle.com/mgornergoogle/getting-started-with-100-flowers-on-tpu/). Do not waste your time for simple data loading and inspecting code\n1. Use Transfer Learning with fine tuning (to avoid over fitting, use small networks such as Inception, DenseNet and EfficientNet) \n    - Simple Ensemble of DenseNet201 and EfficientNetB7 is used\n    - Each network is trained independently then final probabilities are combined together as weighted average \n    - First freeze the weights and trian the added layers, then fine tune the whole network   \n    - Learning rate scheduling is used\n3. Use reasonable data augmentation and visualize samples\n4. Use Label smoothing (makes the model less confident for better generalization)\n\n### Notes: \n- This is not the best code, but a good one for demonstration purposes  \n- EfficientNetB7 is used as in this [code](https://www.kaggle.com/wrrosa/tpu-enet-b7-densenet)\n- Simple weighted average is used to combine both DenseNet201 and EfficientNetB7 (This step could be learnable too) \n\n$ Result = (\\alpha) \\times EfficientNetB7 + (1-\\alpha) \\times DenseNet201 $ \n\n### Versions:\n- V1: uploaded from **0.955**\n- V2: Using both **ENet + DNet**\n- V3: Simple search for alpha to weight **ENet + DNet (0.96+)**\n- V4: Adding notes and next steps Quick Save **(0.96+)**\n- V5: Commit results **(0.96+)**","execution_count":null},{"metadata":{"_uuid":"97e8adbe-39bb-4b53-bdf5-bd4d801df563","_cell_guid":"d9ec7398-6502-4771-a075-22e4bde2bbd6","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":"469d0119-90d2-4c86-ad38-3fea5247524f","_cell_guid":"0e83ccdb-98a5-42de-b0bf-9b5530b9fdbe","trusted":true},"cell_type":"code","source":"import math, re, os\nimport tensorflow as tf\nimport numpy as np\nfrom matplotlib import pyplot as plt\nfrom kaggle_datasets import KaggleDatasets\nfrom sklearn.metrics import f1_score, precision_score, recall_score, confusion_matrix, accuracy_score\nfrom tensorflow.keras import backend as K\nprint(\"Tensorflow version \" + tf.__version__)\nAUTO = tf.data.experimental.AUTOTUNE","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## TPUs","execution_count":null},{"metadata":{"_uuid":"149bff55-6a23-4b07-adc5-9266db6f5097","_cell_guid":"a0f16351-1d7e-4858-b4a9-15cc90849afa","trusted":true},"cell_type":"code","source":"# Detect hardware, return appropriate distribution strategy\ntry:\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":{},"cell_type":"markdown","source":"## Data set","execution_count":null},{"metadata":{"_uuid":"bf84edc7-0a70-43c8-9062-df13d4439f19","_cell_guid":"f1938a7b-b621-4853-ab22-65eba3ab123a","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":{"_uuid":"d37cf704-8d13-4351-963d-64e639c1729d","_cell_guid":"8ac0582b-d7e0-435b-a696-ff712d33e76d","trusted":true},"cell_type":"code","source":"IMAGE_SIZE = [512, 512] # At this size, a GPU will run out of memory. Use the TPU.\n                        # For GPU training, please select 224 x 224 px image size.\nEPOCHS = 20\nBATCH_SIZE = 16 * strategy.num_replicas_in_sync #16 * strategy.num_replicas_in_sync\n\nGCS_PATH_SELECT = { # available image sizes\n    192: GCS_DS_PATH + '/tfrecords-jpeg-192x192',\n    224: GCS_DS_PATH + '/tfrecords-jpeg-224x224',\n    331: GCS_DS_PATH + '/tfrecords-jpeg-331x331',\n    512: GCS_DS_PATH + '/tfrecords-jpeg-512x512'\n}\nGCS_PATH = GCS_PATH_SELECT[IMAGE_SIZE[0]]\n\nTRAINING_FILENAMES = tf.io.gfile.glob(GCS_PATH + '/train/*.tfrec')\nVALIDATION_FILENAMES = tf.io.gfile.glob(GCS_PATH + '/val/*.tfrec')\nTEST_FILENAMES = tf.io.gfile.glob(GCS_PATH + '/test/*.tfrec') # predictions on this dataset should be submitted for the competition\n\nCLASSES = ['pink primrose',    'hard-leaved pocket orchid', 'canterbury bells', 'sweet pea',     'wild geranium',     'tiger lily',           'moon orchid',              'bird of paradise', 'monkshood',        'globe thistle',         # 00 - 09\n           'snapdragon',       \"colt's foot\",               'king protea',      'spear thistle', 'yellow iris',       'globe-flower',         'purple coneflower',        'peruvian lily',    'balloon flower',   'giant white arum lily', # 10 - 19\n           'fire lily',        'pincushion flower',         'fritillary',       'red ginger',    'grape hyacinth',    'corn poppy',           'prince of wales feathers', 'stemless gentian', 'artichoke',        'sweet william',         # 20 - 29\n           'carnation',        'garden phlox',              'love in the mist', 'cosmos',        'alpine sea holly',  'ruby-lipped cattleya', 'cape flower',              'great masterwort', 'siam tulip',       'lenten rose',           # 30 - 39\n           'barberton daisy',  'daffodil',                  'sword lily',       'poinsettia',    'bolero deep blue',  'wallflower',           'marigold',                 'buttercup',        'daisy',            'common dandelion',      # 40 - 49\n           'petunia',          'wild pansy',                'primula',          'sunflower',     'lilac hibiscus',    'bishop of llandaff',   'gaura',                    'geranium',         'orange dahlia',    'pink-yellow dahlia',    # 50 - 59\n           'cautleya spicata', 'japanese anemone',          'black-eyed susan', 'silverbush',    'californian poppy', 'osteospermum',         'spring crocus',            'iris',             'windflower',       'tree poppy',            # 60 - 69\n           'gazania',          'azalea',                    'water lily',       'rose',          'thorn apple',       'morning glory',        'passion flower',           'lotus',            'toad lily',        'anthurium',             # 70 - 79\n           'frangipani',       'clematis',                  'hibiscus',         'columbine',     'desert-rose',       'tree mallow',          'magnolia',                 'cyclamen ',        'watercress',       'canna lily',            # 80 - 89\n           'hippeastrum ',     'bee balm',                  'pink quill',       'foxglove',      'bougainvillea',     'camellia',             'mallow',                   'mexican petunia',  'bromelia',         'blanket flower',        # 90 - 99\n           'trumpet creeper',  'blackberry lily',           'common tulip',     'wild rose']                                                                                                                                               # 100 - 102","execution_count":null,"outputs":[]},{"metadata":{"_uuid":"015caed9-a9a6-4b8b-adbc-94cf86cf74a0","_cell_guid":"adaf831f-e724-4dd2-a566-6188ad55d1bb","trusted":true},"cell_type":"code","source":"# numpy and matplotlib defaults\nnp.set_printoptions(threshold=15, linewidth=80)\n\ndef batch_to_numpy_images_and_labels(data):\n    images, labels = data\n    numpy_images = images.numpy()\n    numpy_labels = labels.numpy()\n    if numpy_labels.dtype == object: # binary string in this case, these are image ID strings\n        numpy_labels = [None for _ in enumerate(numpy_images)]\n    # If no labels, only image IDs, return None for labels (this is the case for test data)\n    return numpy_images, numpy_labels\n\ndef title_from_label_and_target(label, correct_label):\n    if correct_label is None:\n        return CLASSES[label], True\n    correct = (label == correct_label)\n    return \"{} [{}{}{}]\".format(CLASSES[label], 'OK' if correct else 'NO', u\"\\u2192\" if not correct else '',\n                                CLASSES[correct_label] if not correct else ''), correct\n\ndef display_one_flower(image, title, subplot, red=False, titlesize=16):\n    plt.subplot(*subplot)\n    plt.axis('off')\n    plt.imshow(image)\n    if len(title) > 0:\n        plt.title(title, fontsize=int(titlesize) if not red else int(titlesize/1.2), color='red' if red else 'black', fontdict={'verticalalignment':'center'}, pad=int(titlesize/1.5))\n    return (subplot[0], subplot[1], subplot[2]+1)\n    \ndef display_batch_of_images(databatch, predictions=None):\n    \"\"\"This will work with:\n    display_batch_of_images(images)\n    display_batch_of_images(images, predictions)\n    display_batch_of_images((images, labels))\n    display_batch_of_images((images, labels), predictions)\n    \"\"\"\n    # data\n    images, labels = batch_to_numpy_images_and_labels(databatch)\n    if labels is None:\n        labels = [None for _ in enumerate(images)]\n        \n    # auto-squaring: this will drop data that does not fit into square or square-ish rectangle\n    rows = int(math.sqrt(len(images)))\n    cols = len(images)//rows\n        \n    # size and spacing\n    FIGSIZE = 13.0\n    SPACING = 0.1\n    subplot=(rows,cols,1)\n    if rows < cols:\n        plt.figure(figsize=(FIGSIZE,FIGSIZE/cols*rows))\n    else:\n        plt.figure(figsize=(FIGSIZE/rows*cols,FIGSIZE))\n    \n    # display\n    for i, (image, label) in enumerate(zip(images[:rows*cols], labels[:rows*cols])):\n        title = '' if label is None else CLASSES[label]\n        correct = True\n        if predictions is not None:\n            title, correct = title_from_label_and_target(predictions[i], label)\n        dynamic_titlesize = FIGSIZE*SPACING/max(rows,cols)*40+3 # magic formula tested to work from 1x1 to 10x10 images\n        subplot = display_one_flower(image, title, subplot, not correct, titlesize=dynamic_titlesize)\n    \n    #layout\n    plt.tight_layout()\n    if label is None and predictions is None:\n        plt.subplots_adjust(wspace=0, hspace=0)\n    else:\n        plt.subplots_adjust(wspace=SPACING, hspace=SPACING)\n    plt.show()\n\ndef display_confusion_matrix(cmat, score, precision, recall):\n    plt.figure(figsize=(15,15))\n    ax = plt.gca()\n    ax.matshow(cmat, cmap='Reds')\n    ax.set_xticks(range(len(CLASSES)))\n    ax.set_xticklabels(CLASSES, fontdict={'fontsize': 7})\n    plt.setp(ax.get_xticklabels(), rotation=45, ha=\"left\", rotation_mode=\"anchor\")\n    ax.set_yticks(range(len(CLASSES)))\n    ax.set_yticklabels(CLASSES, fontdict={'fontsize': 7})\n    plt.setp(ax.get_yticklabels(), rotation=45, ha=\"right\", rotation_mode=\"anchor\")\n    titlestring = \"\"\n    if score is not None:\n        titlestring += 'f1 = {:.3f} '.format(score)\n    if precision is not None:\n        titlestring += '\\nprecision = {:.3f} '.format(precision)\n    if recall is not None:\n        titlestring += '\\nrecall = {:.3f} '.format(recall)\n    if len(titlestring) > 0:\n        ax.text(101, 1, titlestring, fontdict={'fontsize': 18, 'horizontalalignment':'right', 'verticalalignment':'top', 'color':'#804040'})\n    plt.show()\n    \ndef display_training_curves(training, validation, title, subplot):\n    if subplot%10==1: # set up the subplots on the first call\n        plt.subplots(figsize=(10,10), facecolor='#F0F0F0')\n        plt.tight_layout()\n    ax = plt.subplot(subplot)\n    ax.set_facecolor('#F8F8F8')\n    ax.plot(training)\n    ax.plot(validation)\n    ax.set_title('model '+ title)\n    ax.set_ylabel(title)\n    #ax.set_ylim(0.28,1.05)\n    ax.set_xlabel('epoch')\n    ax.legend(['train', 'valid.'])","execution_count":null,"outputs":[]},{"metadata":{"_uuid":"293cb034-b459-4780-82e4-36b18df7cd4e","_cell_guid":"a3879d3b-df72-4093-bb9e-c630924b4f44","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.one_hot(example['class'], 104)\n    # label = tf.cast(example['class'], tf.int32)  # for custom focal loss\n    \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, num_parallel_reads=AUTO) # 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, num_parallel_calls=AUTO)\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    image = tf.image.random_flip_left_right(image)\n    #image = tf.image.random_saturation(image, 0, 10)\n    #image = tf.image.random_hue(image, 0.05)\n    #image = tf.image.central_crop(image, central_fraction=0.5) #random.uniform(0.7, 0.99)\n    image = tf.image.random_hue(image, 0.05)\n    image = tf.image.random_contrast(image, 0.5, 1.5)\n    image = tf.image.random_brightness(image, 0.25)\n    #image = tf.image.random_contrast(image, 0.6, 2.1)\n    #image = tf.image.random_brightness(image, 0.2)\n    #image = tf.image.resize(image, [512+128, 512+128])\n    #image = tf.image.random_crop(image, size=[512, 512, 3]) # Random crop back\n\n    return image, label   \n\ndef get_training_dataset():\n    dataset = load_dataset(TRAINING_FILENAMES, labeled=True)\n    dataset = dataset.map(data_augment, num_parallel_calls=AUTO)\n    dataset = dataset.repeat() # the training dataset must repeat for several epochs\n    dataset = dataset.shuffle(2048)\n    dataset = dataset.batch(BATCH_SIZE)\n    dataset = dataset.prefetch(AUTO) # prefetch next batch while training (autotune prefetch buffer size)\n    return dataset\n\ndef get_validation_dataset(ordered=False):\n    dataset = load_dataset(VALIDATION_FILENAMES, labeled=True, ordered=ordered)\n    dataset = dataset.batch(BATCH_SIZE)\n    dataset = dataset.cache()\n    dataset = dataset.prefetch(AUTO) # prefetch next batch while training (autotune prefetch buffer size)\n    return dataset\n\ndef get_test_dataset(ordered=False):\n    dataset = load_dataset(TEST_FILENAMES, labeled=False, ordered=ordered)\n    dataset = dataset.batch(BATCH_SIZE)\n    dataset = dataset.prefetch(AUTO) # prefetch next batch while training (autotune prefetch buffer size)\n    return dataset\n\ndef count_data_items(filenames):\n    # the number of data items is written in the name of the .tfrec files, i.e. flowers00-230.tfrec = 230 data items\n    n = [int(re.compile(r\"-([0-9]*)\\.\").search(filename).group(1)) for filename in filenames]\n    return np.sum(n)\n\nNUM_TRAINING_IMAGES = count_data_items(TRAINING_FILENAMES)\nNUM_VALIDATION_IMAGES = count_data_items(VALIDATION_FILENAMES)\nNUM_TEST_IMAGES = count_data_items(TEST_FILENAMES)\nSTEPS_PER_EPOCH = NUM_TRAINING_IMAGES // BATCH_SIZE\nprint('Dataset: {} training images, {} validation images, {} unlabeled test images'.format(NUM_TRAINING_IMAGES, NUM_VALIDATION_IMAGES, NUM_TEST_IMAGES))","execution_count":null,"outputs":[]},{"metadata":{"_uuid":"24509f39-3565-4825-b417-08590c3bcac7","_cell_guid":"15bb2aed-e43b-4540-a8ef-25209b5cdcc8","trusted":true},"cell_type":"code","source":"print(\"Training data shapes:\")\nfor image, label in get_training_dataset().take(3):\n    print(image.numpy().shape, label.numpy().shape)\nprint(\"Training data label examples:\", label.numpy())\nprint(\"Validation data shapes:\")\nfor image, label in get_validation_dataset().take(3):\n    print(image.numpy().shape, label.numpy().shape)\nprint(\"Validation data label examples:\", label.numpy())\nprint(\"Test data shapes:\")\nfor image, idnum in get_test_dataset().take(3):\n    print(image.numpy().shape, idnum.numpy().shape)\nprint(\"Test data IDs:\", idnum.numpy().astype('U')) # U=unicode string","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# augmentation sample\nrow = 5; col = 5;\nall_elements = get_training_dataset().unbatch()\none_element = tf.data.Dataset.from_tensors( next(iter(all_elements)) )\naugmented_element = one_element.repeat().map(data_augment).batch(row*col)\n\nfor (img,label) in augmented_element:\n    plt.figure(figsize=(15,int(15*row/col)))\n    for j in range(row*col):\n        plt.subplot(row,col,j+1)\n        plt.axis('off')\n        plt.imshow(img[j,])\n    plt.show()\n    break","execution_count":null,"outputs":[]},{"metadata":{"_uuid":"1ba32437-d7d9-4133-b9d4-fdf3ce81ea15","_cell_guid":"57f4e543-d97b-4a6b-82fa-98e9371d7f60","trusted":true},"cell_type":"code","source":"# peek at test data\ntest_dataset = get_test_dataset()\ntest_dataset = test_dataset.unbatch().batch(20)\ntest_batch = iter(test_dataset)","execution_count":null,"outputs":[]},{"metadata":{"_uuid":"e448f3d8-227f-4ebf-b40c-0974b2a39c2d","_cell_guid":"dafea472-5766-4b45-8dc9-3454329ebead","trusted":true},"cell_type":"code","source":"# run this cell again for next set of images\ndisplay_batch_of_images(next(test_batch))","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Model Training","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"# Learning rate schedule\n# Using an LR ramp up because fine-tuning a pre-trained model.\n# Starting with a high LR would break the pre-trained weights.\n\nLR_START = 0.00001\nLR_MAX = 0.00005 * strategy.num_replicas_in_sync\nLR_MIN = 0.00001\nLR_RAMPUP_EPOCHS = 5\nLR_SUSTAIN_EPOCHS = 0\nLR_EXP_DECAY = .8\n\ndef lrfn(epoch):\n    if epoch < LR_RAMPUP_EPOCHS:\n        lr = (LR_MAX - LR_START) / LR_RAMPUP_EPOCHS * epoch + LR_START\n    elif epoch < LR_RAMPUP_EPOCHS + LR_SUSTAIN_EPOCHS:\n        lr = LR_MAX\n    else:\n        lr = (LR_MAX - LR_MIN) * LR_EXP_DECAY**(epoch - LR_RAMPUP_EPOCHS - LR_SUSTAIN_EPOCHS) + LR_MIN\n    return lr\n    \nlr_callback = tf.keras.callbacks.LearningRateScheduler(lrfn, verbose=True)\n\nrng = [i for i in range(EPOCHS)]\ny = [lrfn(x) for x in rng]\nplt.plot(rng, y)\nprint(\"Learning rate schedule: {:.3g} to {:.3g} to {:.3g}\".format(y[0], max(y), y[-1]))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# change the label_smoothing value (0.1 for example) and compare results  \ndef categorical_smooth_loss(y_true, y_pred, label_smoothing=0.1):\n    loss = tf.keras.losses.categorical_crossentropy(y_true, y_pred, label_smoothing=label_smoothing)\n    return loss","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### ENet: Freeze the pre trained model and train the added layers ","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"!pip install -q efficientnet","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"import efficientnet.tfkeras as efn","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"with strategy.scope():\n    enet = efn.EfficientNetB7(input_shape=(512, 512, 3), weights='noisy-student', include_top=False)\n    enet.trainable = False \n    \n    model1 = tf.keras.Sequential([\n        enet,\n        tf.keras.layers.GlobalAveragePooling2D(),\n        tf.keras.layers.Dense(len(CLASSES), activation='softmax')\n    ])\n        \nmodel1.compile(\n    optimizer=tf.keras.optimizers.Adam(lr=0.0001),\n    loss = categorical_smooth_loss,\n    metrics=['categorical_accuracy']\n)\nmodel1.summary()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"model1.fit(get_training_dataset(), steps_per_epoch=STEPS_PER_EPOCH, epochs=5, validation_data=get_validation_dataset()) ","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### ENet: Fine-tune the whole model","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"enet.trainable = True\nmodel1.compile(\n    optimizer='adam',\n    loss = categorical_smooth_loss,\n    metrics=['categorical_accuracy']\n)\nmodel1.summary()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"history1 = model1.fit(get_training_dataset(), steps_per_epoch=STEPS_PER_EPOCH, epochs=EPOCHS, callbacks = [lr_callback], validation_data=get_validation_dataset())","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"display_training_curves(history1.history['loss'], history1.history['val_loss'], 'loss', 211)\ndisplay_training_curves(history1.history['categorical_accuracy'], history1.history['val_categorical_accuracy'], 'accuracy', 212)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### DensNet: Freeze the pre trained model and train the added layers ","execution_count":null},{"metadata":{"_uuid":"8d77a600-80f1-4455-bf1e-eba3ccf11f83","_cell_guid":"9c995751-3275-4850-9176-431392a5f3d2","trusted":true},"cell_type":"code","source":"with strategy.scope():\n\n    dnet = tf.keras.applications.DenseNet201(weights='imagenet', include_top=False ,input_shape=[*IMAGE_SIZE, 3])\n    dnet.trainable = False \n    \n    model2 = tf.keras.Sequential([\n        dnet,\n        tf.keras.layers.GlobalAveragePooling2D(),\n        tf.keras.layers.Dense(len(CLASSES), activation='softmax')\n    ])\n\n    \nmodel2.compile(\n    optimizer='adam',\n    loss = categorical_smooth_loss,\n    metrics=['categorical_accuracy']\n)\nmodel2.summary()","execution_count":null,"outputs":[]},{"metadata":{"_uuid":"fbcb4922-b7d2-4b9a-bc96-18932658e6aa","_cell_guid":"9e073bfa-b7c0-4fb5-a23b-29e20cc830e1","trusted":true},"cell_type":"code","source":"model2.fit(get_training_dataset(), steps_per_epoch=STEPS_PER_EPOCH, epochs=5, validation_data=get_validation_dataset())","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### DenseNet: Fine-tune the whole model ","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"dnet.trainable = True\nmodel2.compile(\n    optimizer='adam',\n    loss = categorical_smooth_loss,\n    metrics=['categorical_accuracy']\n)\nmodel2.summary()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"history2 = model2.fit(get_training_dataset(), steps_per_epoch=STEPS_PER_EPOCH, epochs=EPOCHS, callbacks = [lr_callback], validation_data=get_validation_dataset())","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"display_training_curves(history2.history['loss'], history2.history['val_loss'], 'loss', 211)\ndisplay_training_curves(history2.history['categorical_accuracy'], history2.history['val_categorical_accuracy'], 'accuracy', 212)","execution_count":null,"outputs":[]},{"metadata":{"_uuid":"2a70238b-eeed-4cfd-8ae3-8eca6faf19f3","_cell_guid":"68eb1c95-a32b-4068-b3ef-dd2a19ab3ece","trusted":true},"cell_type":"code","source":"cmdataset = get_validation_dataset(ordered=True) # since we are splitting the dataset and iterating separately on images and labels, order matters.\nimages_ds = cmdataset.map(lambda image, label: image)\nlabels_ds = cmdataset.map(lambda image, label: label).unbatch()\ncm_correct_labels = next(iter(labels_ds.batch(NUM_VALIDATION_IMAGES))).numpy() # get everything as one batch\n\ncm_correct_labels = np.argmax(cm_correct_labels, axis=-1)\n\ncm_probabilities1 = model1.predict(images_ds)\ncm_probabilities2 = model2.predict(images_ds)\n\n\n# find best weight between models (can be another model!)\nscores = []\nfor alpha in np.linspace(0,1,100):\n    cm_probabilities = alpha*cm_probabilities1+(1-alpha)*cm_probabilities2\n    cm_predictions = np.argmax(cm_probabilities, axis=-1)\n    scores.append(accuracy_score(cm_correct_labels, cm_predictions))\n\nbest_alpha = np.argmax(scores)/100\nplt.title(\"(alpha)ENET + (1-alpha)DNET; Best alpha = \" + str(best_alpha))\nplt.ylabel(\"Accuracy\")\nplt.xlabel(\"alpha\")\nplt.plot(scores)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Results","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"cm_probabilities = best_alpha*cm_probabilities1+(1-best_alpha)*cm_probabilities2\ncm_predictions = np.argmax(cm_probabilities, axis=-1)\n\nprint(\"Correct   labels: \", cm_correct_labels.shape, cm_correct_labels)\nprint(\"Predicted labels: \", cm_predictions.shape, cm_predictions)","execution_count":null,"outputs":[]},{"metadata":{"_uuid":"c68ae9c8-1d61-4b22-b0f9-d2caf8dfe5a9","_cell_guid":"3bf197d8-0288-4fa5-b7b5-053e59bc9624","trusted":true},"cell_type":"code","source":"cmat = confusion_matrix(cm_correct_labels, cm_predictions, labels=range(len(CLASSES)))\nscore = f1_score(cm_correct_labels, cm_predictions, labels=range(len(CLASSES)), average='macro')\nprecision = precision_score(cm_correct_labels, cm_predictions, labels=range(len(CLASSES)), average='macro')\nrecall = recall_score(cm_correct_labels, cm_predictions, labels=range(len(CLASSES)), average='macro')\ncmat = (cmat.T / cmat.sum(axis=1)).T # normalized\ndisplay_confusion_matrix(cmat, score, precision, recall)\nprint('f1 score: {:.3f}, precision: {:.3f}, recall: {:.3f}'.format(score, precision, recall))","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Save submission.csv","execution_count":null},{"metadata":{"_uuid":"e7940829-0c56-4438-b421-ab298917e6a9","_cell_guid":"3a476ea8-e03e-46c2-9bf8-697d7bb18175","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)\n#probabilities = model.predict(test_images_ds)\n\nprobabilities = best_alpha*model1.predict(test_images_ds) + (1-best_alpha)*model2.predict(test_images_ds)\n\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='')\nprint(\"submission.csv is saved\")\n","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### Next Steps:\n- Use more Networks (EfficientNet)\n- More data augmentation (Rotation with small angles)\n- Try Focal Loss with different settings \n- Try label smoothing with different settings \n- Train Ensemble as one network, end to end","execution_count":null}],"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}