{"cells":[{"metadata":{"trusted":true},"cell_type":"code","source":"%matplotlib inline\nfrom IPython.core.interactiveshell import InteractiveShell\nInteractiveShell.ast_node_interactivity = \"all\"\nimport os\nimport json\nimport datetime as dt\nimport matplotlib.pyplot as plt\nplt.rcParams['figure.figsize'] = [16, 10]\nplt.rcParams['font.size'] = 14\nimport seaborn as sns\nimport cv2\nimport pandas as pd\nimport numpy as np\nimport tensorflow as tf\nfrom tensorflow import keras\nfrom tensorflow.keras.layers import Conv2D, MaxPooling2D\nfrom tensorflow.keras.layers import Dense, Dropout, Flatten, Activation\nfrom tensorflow.keras.metrics import categorical_accuracy, top_k_categorical_accuracy, categorical_crossentropy\nfrom tensorflow.keras.models import Sequential\nfrom tensorflow.keras.callbacks import EarlyStopping, ReduceLROnPlateau, ModelCheckpoint\nfrom tensorflow.keras.optimizers import Adam\nfrom tensorflow.keras.applications import MobileNet\nfrom tensorflow.keras.applications.mobilenet import preprocess_input\nstart = dt.datetime.now()","execution_count":49,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"DP_DIR = '../input/shuffle-csvs/'\nINPUT_DIR = '../input/quickdraw-doodle-recognition/'\n\nBASE_SIZE = 256\nNCSVS = 100\nNCATS = 340\nnp.random.seed(seed=1987)\ntf.set_random_seed(seed=1987)\n\ndef f2cat(filename: str) -> str:\n    return filename.split('.')[0]\n\ndef list_all_categories():\n    files = os.listdir(os.path.join(INPUT_DIR, 'train_simplified'))\n    return sorted([f2cat(f) for f in files], key=str.lower)","execution_count":50,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def apk(actual, predicted, k=3):\n    \"\"\"\n    Source: https://github.com/benhamner/Metrics/blob/master/Python/ml_metrics/average_precision.py\n    \"\"\"\n    if len(predicted) > k:\n        predicted = predicted[:k]\n    score = 0.0\n    num_hits = 0.0\n    for i, p in enumerate(predicted):\n        if p in actual and p not in predicted[:i]:\n            num_hits += 1.0\n            score += num_hits / (i + 1.0)\n    if not actual:\n        return 0.0\n    return score / min(len(actual), k)\n\ndef mapk(actual, predicted, k=3):\n    \"\"\"\n    Source: https://github.com/benhamner/Metrics/blob/master/Python/ml_metrics/average_precision.py\n    \"\"\"\n    return np.mean([apk(a, p, k) for a, p in zip(actual, predicted)])\n\ndef preds2catids(predictions):\n    return pd.DataFrame(np.argsort(-predictions, axis=1)[:, :3], columns=['a', 'b', 'c'])\n\ndef top_3_accuracy(y_true, y_pred):\n    return top_k_categorical_accuracy(y_true, y_pred, k=3)","execution_count":51,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"STEPS = 800\nEPOCHS = 16\nsize = 64\nbatchsize = 680","execution_count":52,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"model = MobileNet(input_shape=(size, size, 1), alpha=1., weights=None, classes=NCATS)\nmodel.compile(optimizer=Adam(lr=0.002), loss='categorical_crossentropy',\n              metrics=[categorical_crossentropy, categorical_accuracy, top_3_accuracy])\nprint(model.summary())","execution_count":53,"outputs":[{"output_type":"stream","text":"WARNING:tensorflow:From /opt/conda/lib/python3.6/site-packages/tensorflow/python/ops/resource_variable_ops.py:435: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\nInstructions for updating:\nColocations handled automatically by placer.\nWARNING:tensorflow:From /opt/conda/lib/python3.6/site-packages/tensorflow/python/keras/layers/core.py:143: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.\nInstructions for updating:\nPlease use `rate` instead of `keep_prob`. Rate should be set to `rate = 1 - keep_prob`.\n_________________________________________________________________\nLayer (type)                 Output Shape              Param #   \n=================================================================\ninput_1 (InputLayer)         (None, 64, 64, 1)         0         \n_________________________________________________________________\nconv1_pad (ZeroPadding2D)    (None, 65, 65, 1)         0         \n_________________________________________________________________\nconv1 (Conv2D)               (None, 32, 32, 32)        288       \n_________________________________________________________________\nconv1_bn (BatchNormalization (None, 32, 32, 32)        128       \n_________________________________________________________________\nconv1_relu (ReLU)            (None, 32, 32, 32)        0         \n_________________________________________________________________\nconv_dw_1 (DepthwiseConv2D)  (None, 32, 32, 32)        288       \n_________________________________________________________________\nconv_dw_1_bn (BatchNormaliza (None, 32, 32, 32)        128       \n_________________________________________________________________\nconv_dw_1_relu (ReLU)        (None, 32, 32, 32)        0         \n_________________________________________________________________\nconv_pw_1 (Conv2D)           (None, 32, 32, 64)        2048      \n_________________________________________________________________\nconv_pw_1_bn (BatchNormaliza (None, 32, 32, 64)        256       \n_________________________________________________________________\nconv_pw_1_relu (ReLU)        (None, 32, 32, 64)        0         \n_________________________________________________________________\nconv_pad_2 (ZeroPadding2D)   (None, 33, 33, 64)        0         \n_________________________________________________________________\nconv_dw_2 (DepthwiseConv2D)  (None, 16, 16, 64)        576       \n_________________________________________________________________\nconv_dw_2_bn (BatchNormaliza (None, 16, 16, 64)        256       \n_________________________________________________________________\nconv_dw_2_relu (ReLU)        (None, 16, 16, 64)        0         \n_________________________________________________________________\nconv_pw_2 (Conv2D)           (None, 16, 16, 128)       8192      \n_________________________________________________________________\nconv_pw_2_bn (BatchNormaliza (None, 16, 16, 128)       512       \n_________________________________________________________________\nconv_pw_2_relu (ReLU)        (None, 16, 16, 128)       0         \n_________________________________________________________________\nconv_dw_3 (DepthwiseConv2D)  (None, 16, 16, 128)       1152      \n_________________________________________________________________\nconv_dw_3_bn (BatchNormaliza (None, 16, 16, 128)       512       \n_________________________________________________________________\nconv_dw_3_relu (ReLU)        (None, 16, 16, 128)       0         \n_________________________________________________________________\nconv_pw_3 (Conv2D)           (None, 16, 16, 128)       16384     \n_________________________________________________________________\nconv_pw_3_bn (BatchNormaliza (None, 16, 16, 128)       512       \n_________________________________________________________________\nconv_pw_3_relu (ReLU)        (None, 16, 16, 128)       0         \n_________________________________________________________________\nconv_pad_4 (ZeroPadding2D)   (None, 17, 17, 128)       0         \n_________________________________________________________________\nconv_dw_4 (DepthwiseConv2D)  (None, 8, 8, 128)         1152      \n_________________________________________________________________\nconv_dw_4_bn (BatchNormaliza (None, 8, 8, 128)         512       \n_________________________________________________________________\nconv_dw_4_relu (ReLU)        (None, 8, 8, 128)         0         \n_________________________________________________________________\nconv_pw_4 (Conv2D)           (None, 8, 8, 256)         32768     \n_________________________________________________________________\nconv_pw_4_bn (BatchNormaliza (None, 8, 8, 256)         1024      \n_________________________________________________________________\nconv_pw_4_relu (ReLU)        (None, 8, 8, 256)         0         \n_________________________________________________________________\nconv_dw_5 (DepthwiseConv2D)  (None, 8, 8, 256)         2304      \n_________________________________________________________________\nconv_dw_5_bn (BatchNormaliza (None, 8, 8, 256)         1024      \n_________________________________________________________________\nconv_dw_5_relu (ReLU)        (None, 8, 8, 256)         0         \n_________________________________________________________________\nconv_pw_5 (Conv2D)           (None, 8, 8, 256)         65536     \n_________________________________________________________________\nconv_pw_5_bn (BatchNormaliza (None, 8, 8, 256)         1024      \n_________________________________________________________________\nconv_pw_5_relu (ReLU)        (None, 8, 8, 256)         0         \n_________________________________________________________________\nconv_pad_6 (ZeroPadding2D)   (None, 9, 9, 256)         0         \n_________________________________________________________________\nconv_dw_6 (DepthwiseConv2D)  (None, 4, 4, 256)         2304      \n_________________________________________________________________\nconv_dw_6_bn (BatchNormaliza (None, 4, 4, 256)         1024      \n_________________________________________________________________\nconv_dw_6_relu (ReLU)        (None, 4, 4, 256)         0         \n_________________________________________________________________\nconv_pw_6 (Conv2D)           (None, 4, 4, 512)         131072    \n_________________________________________________________________\nconv_pw_6_bn (BatchNormaliza (None, 4, 4, 512)         2048      \n_________________________________________________________________\nconv_pw_6_relu (ReLU)        (None, 4, 4, 512)         0         \n_________________________________________________________________\nconv_dw_7 (DepthwiseConv2D)  (None, 4, 4, 512)         4608      \n_________________________________________________________________\nconv_dw_7_bn (BatchNormaliza (None, 4, 4, 512)         2048      \n_________________________________________________________________\nconv_dw_7_relu (ReLU)        (None, 4, 4, 512)         0         \n_________________________________________________________________\nconv_pw_7 (Conv2D)           (None, 4, 4, 512)         262144    \n_________________________________________________________________\nconv_pw_7_bn (BatchNormaliza (None, 4, 4, 512)         2048      \n_________________________________________________________________\nconv_pw_7_relu (ReLU)        (None, 4, 4, 512)         0         \n_________________________________________________________________\nconv_dw_8 (DepthwiseConv2D)  (None, 4, 4, 512)         4608      \n_________________________________________________________________\nconv_dw_8_bn (BatchNormaliza (None, 4, 4, 512)         2048      \n_________________________________________________________________\nconv_dw_8_relu (ReLU)        (None, 4, 4, 512)         0         \n_________________________________________________________________\nconv_pw_8 (Conv2D)           (None, 4, 4, 512)         262144    \n_________________________________________________________________\nconv_pw_8_bn (BatchNormaliza (None, 4, 4, 512)         2048      \n_________________________________________________________________\nconv_pw_8_relu (ReLU)        (None, 4, 4, 512)         0         \n_________________________________________________________________\nconv_dw_9 (DepthwiseConv2D)  (None, 4, 4, 512)         4608      \n_________________________________________________________________\nconv_dw_9_bn (BatchNormaliza (None, 4, 4, 512)         2048      \n_________________________________________________________________\nconv_dw_9_relu (ReLU)        (None, 4, 4, 512)         0         \n_________________________________________________________________\nconv_pw_9 (Conv2D)           (None, 4, 4, 512)         262144    \n_________________________________________________________________\nconv_pw_9_bn (BatchNormaliza (None, 4, 4, 512)         2048      \n_________________________________________________________________\nconv_pw_9_relu (ReLU)        (None, 4, 4, 512)         0         \n_________________________________________________________________\nconv_dw_10 (DepthwiseConv2D) (None, 4, 4, 512)         4608      \n_________________________________________________________________\nconv_dw_10_bn (BatchNormaliz (None, 4, 4, 512)         2048      \n_________________________________________________________________\nconv_dw_10_relu (ReLU)       (None, 4, 4, 512)         0         \n_________________________________________________________________\nconv_pw_10 (Conv2D)          (None, 4, 4, 512)         262144    \n_________________________________________________________________\nconv_pw_10_bn (BatchNormaliz (None, 4, 4, 512)         2048      \n_________________________________________________________________\nconv_pw_10_relu (ReLU)       (None, 4, 4, 512)         0         \n_________________________________________________________________\nconv_dw_11 (DepthwiseConv2D) (None, 4, 4, 512)         4608      \n_________________________________________________________________\nconv_dw_11_bn (BatchNormaliz (None, 4, 4, 512)         2048      \n_________________________________________________________________\nconv_dw_11_relu (ReLU)       (None, 4, 4, 512)         0         \n_________________________________________________________________\nconv_pw_11 (Conv2D)          (None, 4, 4, 512)         262144    \n_________________________________________________________________\nconv_pw_11_bn (BatchNormaliz (None, 4, 4, 512)         2048      \n_________________________________________________________________\nconv_pw_11_relu (ReLU)       (None, 4, 4, 512)         0         \n_________________________________________________________________\nconv_pad_12 (ZeroPadding2D)  (None, 5, 5, 512)         0         \n_________________________________________________________________\nconv_dw_12 (DepthwiseConv2D) (None, 2, 2, 512)         4608      \n_________________________________________________________________\nconv_dw_12_bn (BatchNormaliz (None, 2, 2, 512)         2048      \n_________________________________________________________________\nconv_dw_12_relu (ReLU)       (None, 2, 2, 512)         0         \n_________________________________________________________________\nconv_pw_12 (Conv2D)          (None, 2, 2, 1024)        524288    \n_________________________________________________________________\nconv_pw_12_bn (BatchNormaliz (None, 2, 2, 1024)        4096      \n_________________________________________________________________\nconv_pw_12_relu (ReLU)       (None, 2, 2, 1024)        0         \n_________________________________________________________________\nconv_dw_13 (DepthwiseConv2D) (None, 2, 2, 1024)        9216      \n_________________________________________________________________\nconv_dw_13_bn (BatchNormaliz (None, 2, 2, 1024)        4096      \n_________________________________________________________________\nconv_dw_13_relu (ReLU)       (None, 2, 2, 1024)        0         \n_________________________________________________________________\nconv_pw_13 (Conv2D)          (None, 2, 2, 1024)        1048576   \n_________________________________________________________________\nconv_pw_13_bn (BatchNormaliz (None, 2, 2, 1024)        4096      \n_________________________________________________________________\nconv_pw_13_relu (ReLU)       (None, 2, 2, 1024)        0         \n_________________________________________________________________\nglobal_average_pooling2d (Gl (None, 1024)              0         \n_________________________________________________________________\nreshape_1 (Reshape)          (None, 1, 1, 1024)        0         \n_________________________________________________________________\ndropout (Dropout)            (None, 1, 1, 1024)        0         \n_________________________________________________________________\nconv_preds (Conv2D)          (None, 1, 1, 340)         348500    \n_________________________________________________________________\nact_softmax (Activation)     (None, 1, 1, 340)         0         \n_________________________________________________________________\nreshape_2 (Reshape)          (None, 340)               0         \n=================================================================\nTotal params: 3,576,788\nTrainable params: 3,554,900\nNon-trainable params: 21,888\n_________________________________________________________________\nNone\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"def draw_cv2(raw_strokes, size=256, lw=6, time_color=True):\n    img = np.zeros((BASE_SIZE, BASE_SIZE), np.uint8)\n    for t, stroke in enumerate(raw_strokes):\n        for i in range(len(stroke[0]) - 1):\n            color = 255 - min(t, 10) * 13 if time_color else 255\n            _ = cv2.line(img, (stroke[0][i], stroke[1][i]),\n                         (stroke[0][i + 1], stroke[1][i + 1]), color, lw)\n    if size != BASE_SIZE:\n        return cv2.resize(img, (size, size))\n    else:\n        return img\n\ndef image_generator_xd(size, batchsize, ks, lw=6, time_color=True):\n    while True:\n        for k in np.random.permutation(ks):\n            filename = os.path.join(DP_DIR, 'train_k{}.csv.gz'.format(k))\n            for df in pd.read_csv(filename, chunksize=batchsize):\n                df['drawing'] = df['drawing'].apply(json.loads)\n                x = np.zeros((len(df), size, size, 1))\n                for i, raw_strokes in enumerate(df.drawing.values):\n                    x[i, :, :, 0] = draw_cv2(raw_strokes, size=size, lw=lw,\n                                             time_color=time_color)\n                x = preprocess_input(x).astype(np.float32)\n                y = keras.utils.to_categorical(df.y, num_classes=NCATS)\n                yield x, y\n\ndef df_to_image_array_xd(df, size, lw=6, time_color=True):\n    df['drawing'] = df['drawing'].apply(json.loads)\n    x = np.zeros((len(df), size, size, 1))\n    for i, raw_strokes in enumerate(df.drawing.values):\n        x[i, :, :, 0] = draw_cv2(raw_strokes, size=size, lw=lw, time_color=time_color)\n    x = preprocess_input(x).astype(np.float32)\n    return x","execution_count":54,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"valid_df = pd.read_csv(os.path.join(DP_DIR, 'train_k{}.csv.gz'.format(NCSVS - 1)), nrows=34000)\nx_valid = df_to_image_array_xd(valid_df, size)\ny_valid = keras.utils.to_categorical(valid_df.y, num_classes=NCATS)\nprint(x_valid.shape, y_valid.shape)\nprint('Validation array memory {:.2f} GB'.format(x_valid.nbytes / 1024.**3 ))","execution_count":55,"outputs":[{"output_type":"stream","text":"(34000, 64, 64, 1) (34000, 340)\nValidation array memory 0.52 GB\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_datagen = image_generator_xd(size=size, batchsize=batchsize, ks=range(NCSVS - 1))","execution_count":56,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"x, y = next(train_datagen)\nn = 8\nfig, axs = plt.subplots(nrows=n, ncols=n, sharex=True, sharey=True, figsize=(12, 12))\nfor i in range(n**2):\n    ax = axs[i // n, i % n]\n    (-x[i]+1)/2\n    ax.imshow((-x[i, :, :, 0] + 1)/2, cmap=plt.cm.gray)\n    ax.axis('off')\nplt.tight_layout()\nfig.savefig('gs.png', dpi=300)\nplt.show();\n","execution_count":57,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 864x864 with 64 Axes>","image/png":"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\n"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"%%timeit\nx, y = next(train_datagen)","execution_count":58,"outputs":[{"output_type":"stream","text":"109 ms ± 292 µs per loop (mean ± std. dev. of 7 runs, 10 loops each)\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"callbacks = [\n    ReduceLROnPlateau(monitor='val_top_3_accuracy', factor=0.75, patience=3, min_delta=0.001,\n                          mode='max', min_lr=1e-5, verbose=1),\n    ModelCheckpoint('model.h5', monitor='val_top_3_accuracy', mode='max', save_best_only=True,\n                    save_weights_only=True),\n]\nhists = []\nhist = model.fit_generator(\n    train_datagen, steps_per_epoch=STEPS, epochs=70, verbose=1,\n    validation_data=(x_valid, y_valid),\n    callbacks = callbacks\n)\nhists.append(hist)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"hist_df = pd.concat([pd.DataFrame(hist.history) for hist in hists], sort=True)\nhist_df.index = np.arange(1, len(hist_df)+1)\nfig, axs = plt.subplots(nrows=2, sharex=True, figsize=(16, 10))\naxs[0].plot(hist_df.val_categorical_accuracy, lw=5, label='Validation Accuracy')\naxs[0].plot(hist_df.categorical_accuracy, lw=5, label='Training Accuracy')\naxs[0].set_ylabel('Accuracy')\naxs[0].set_xlabel('Epoch')\naxs[0].grid()\naxs[0].legend(loc=0)\naxs[1].plot(hist_df.val_categorical_crossentropy, lw=5, label='Validation MLogLoss')\naxs[1].plot(hist_df.categorical_crossentropy, lw=5, label='Training MLogLoss')\naxs[1].set_ylabel('MLogLoss')\naxs[1].set_xlabel('Epoch')\naxs[1].grid()\naxs[1].legend(loc=0)\nfig.savefig('hist.png', dpi=300)\nplt.show();","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"valid_predictions = model.predict(x_valid, batch_size=128, verbose=1)\nmap3 = mapk(valid_df[['y']].values, preds2catids(valid_predictions).values)\nprint('Map3: {:.3f}'.format(map3))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"test = pd.read_csv(os.path.join(INPUT_DIR, 'test_simplified.csv'))\ntest.head()\nx_test = df_to_image_array_xd(test, size)\nprint(test.shape, x_test.shape)\nprint('Test array memory {:.2f} GB'.format(x_test.nbytes / 1024.**3 ))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"test_predictions = model.predict(x_test, batch_size=128, verbose=1)\n\ntop3 = preds2catids(test_predictions)\ntop3.head()\ntop3.shape\n\ncats = list_all_categories()\nid2cat = {k: cat.replace(' ', '_') for k, cat in enumerate(cats)}\ntop3cats = top3.replace(id2cat)\ntop3cats.head()\ntop3cats.shape","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"test['word'] = top3cats['a'] + ' ' + top3cats['b'] + ' ' + top3cats['c']\nsubmission = test[['key_id', 'word']]\nsubmission.to_csv('gs_mn_submission_{}.csv'.format(int(map3 * 10**4)), index=False)\nsubmission.head()\nsubmission.shape","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"end = dt.datetime.now()\nprint('Latest run {}.\\nTotal time {}s'.format(end, (end - start).seconds))","execution_count":null,"outputs":[]}],"metadata":{"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"name":"python","version":"3.6.4","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"}},"nbformat":4,"nbformat_minor":1}