{"cells":[{"metadata":{"trusted":true},"cell_type":"code","source":"import os","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"import numpy as np\nimport pandas as pd\n\nimport matplotlib.pyplot as plt\n%matplotlib inline","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"import tensorflow as tf\nfrom tensorflow.keras.layers import Dense, Input, Embedding, Flatten, Dropout, LSTM, Bidirectional\nfrom tensorflow.keras.optimizers import Adam\nfrom tensorflow.keras.models import Model, Sequential\nfrom tensorflow.keras.callbacks import ModelCheckpoint","execution_count":null,"outputs":[]},{"metadata":{"_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","trusted":true},"cell_type":"code","source":"import transformers\n\nfrom tokenizers import BertWordPieceTokenizer\n\nfrom tqdm.notebook import tqdm\n\nfrom kaggle_datasets import KaggleDatasets","execution_count":null,"outputs":[]},{"metadata":{"_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","trusted":true},"cell_type":"code","source":"def fast_encode(texts, tokenizer, chunk_size=256, maxlen=512):\n    \"\"\"\n    Tokenize text\n    Source: https://www.kaggle.com/xhlulu/jigsaw-tpu-distilbert-with-huggingface-and-keras\n    \"\"\"\n    tokenizer.enable_truncation(max_length=maxlen)\n    tokenizer.enable_padding(max_length=maxlen)\n    all_ids = []\n    \n    for i in tqdm(range(0, len(texts), chunk_size)):\n        text_chunk = texts[i:i+chunk_size].tolist()\n        encs = tokenizer.encode_batch(text_chunk)\n        all_ids.extend([enc.ids for enc in encs])\n    \n    return np.array(all_ids)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def build_model(transformer, max_len=512):\n    \"\"\"\n    Model initalization\n    Source: https://www.kaggle.com/xhlulu/jigsaw-tpu-distilbert-with-huggingface-and-keras\n    \"\"\"\n    input_word_ids = Input(shape=(max_len,), dtype=tf.int32, name=\"input_word_ids\")\n    sequence_output = transformer(input_word_ids)[0]\n    cls_token = sequence_output[:, 0, :]\n    dense_layer = Dense(224, activation='relu')(cls_token)\n    dense_layer = Dropout(0.2)(dense_layer)\n    out = Dense(224, activation='relu')(dense_layer)\n    out = Dense(1, activation='sigmoid')(out)\n    model = Model(inputs=input_word_ids, outputs=out)\n    # model = InceptionV3(input_tensor=input_word_ids, weights='imagenet', include_top=True)\n    model.compile(Adam(lr=1e-3), loss='binary_crossentropy', metrics=['accuracy'])\n    \n    return model\n\n# def build_model(transformer, max_len=512):\n#     model = Sequential()\n#     model.add(Embedding(119547, 500, input_length=MAX_LEN))\n#     model.add(Bidirectional(LSTM(300, dropout=0.3, recurrent_dropout=0.3)))\n#     model.add(Dense(256, activation='relu'))\n#     model.add(Dropout(0.2))\n#     model.add(Dense(1, activation='sigmoid'))\n#     model.compile(Adam(lr=1e-5), loss='binary_crossentropy', metrics=['accuracy'])\n","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"\n# TPU Configs","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"# Detect hardware, return appropriate distribution strategy\ntry:\n    # TPU detection. No parameters necessary if TPU_NAME environment variable is\n    # set: this is always the case on Kaggle.\n    tpu = tf.distribute.cluster_resolver.TPUClusterResolver()\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    # Default distribution strategy in Tensorflow. Works on CPU and single GPU.\n    strategy = tf.distribute.get_strategy()\n\nprint(\"REPLICAS: \", strategy.num_replicas_in_sync)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"AUTO = tf.data.experimental.AUTOTUNE\n\n# Data access\n# GCS_DS_PATH = KaggleDatasets().get_gcs_path()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# Configuration\nEPOCHS = 2\nBATCH_SIZE = 64 * strategy.num_replicas_in_sync\nMAX_LEN = 200 #192","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Create fast tokenizer","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"tokenizer = transformers.DistilBertTokenizer.from_pretrained('distilbert-base-multilingual-cased')\ntokenizer.save_pretrained('.')\nfast_tokenizer = BertWordPieceTokenizer('vocab.txt', lowercase=True)\nfast_tokenizer","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Load text data","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"DATA_PATH = \"/kaggle/input/jigsaw-multilingual-toxic-comment-classification/\"","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train1 = pd.read_csv(os.path.join(DATA_PATH, \"jigsaw-toxic-comment-train.csv\"))\ntrain2 = pd.read_csv(os.path.join(DATA_PATH, \"jigsaw-unintended-bias-train.csv\"))\ntrain2.toxic = train2.toxic.round().astype(int)\n\nvalid = pd.read_csv(os.path.join(DATA_PATH, 'validation.csv'))\ntest = pd.read_csv(os.path.join(DATA_PATH, 'test.csv'))\nsub = pd.read_csv(os.path.join(DATA_PATH, 'sample_submission.csv'))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train1.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# Combine train1 with a subset of train2\ntrain = pd.concat([\n    train1[['comment_text', 'toxic']],\n    train2[['comment_text', 'toxic']].query('toxic==1'),\n    \n])\n\n# Note: changed random_state from 0 to 39","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train.toxic.value_counts()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"x_train = fast_encode(train.comment_text.astype(str), fast_tokenizer, maxlen=MAX_LEN)\nx_valid = fast_encode(valid.comment_text.astype(str), fast_tokenizer, maxlen=MAX_LEN)\nx_test = fast_encode(test.content.astype(str), fast_tokenizer, maxlen=MAX_LEN)\n\ny_train = train.toxic.values\ny_valid = valid.toxic.values","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Build datasets objects","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"train_dataset = (\n    tf.data.Dataset\n    .from_tensor_slices((x_train, y_train))\n    .repeat()\n    .shuffle(2048)\n    .batch(BATCH_SIZE)\n    .prefetch(AUTO)\n)\n\nvalid_dataset = (\n    tf.data.Dataset\n    .from_tensor_slices((x_valid, y_valid))\n    .batch(BATCH_SIZE)\n    .cache()\n    .prefetch(AUTO)\n)\n\ntest_dataset = (\n    tf.data.Dataset\n    .from_tensor_slices(x_test)\n    .batch(BATCH_SIZE)\n)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Load model into the TPU","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"%%time\nwith strategy.scope():\n    transformer_layer = (\n        transformers.TFDistilBertModel\n        .from_pretrained('distilbert-base-multilingual-cased')\n    )\n    model = build_model(transformer_layer, max_len=MAX_LEN)\nmodel.summary()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Train Model","execution_count":null},{"metadata":{},"cell_type":"markdown","source":"First, we train on the subset of the training set, which is completely in English.","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"n_steps = x_train.shape[0] // BATCH_SIZE\ntrain_history = model.fit(\n    train_dataset,\n    steps_per_epoch=n_steps,\n    validation_data=valid_dataset,\n    epochs=1#2\n)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"Now that we have pretty much saturated the learning potential of the model on english only data, we train it for one more epoch on the `validation` set, which is significantly smaller but contains a mixture of different languages.","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Submission","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"# sub = pd.read_csv(os.path.join(DATA_PATH, 'sample_submission.csv'))\n# # print(submi.shape)\n# sub = model.predict(test_dataset, verbose=1)\n# # sub['toxic1'].to_csv('submission.csv', index=True)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# sub\n# submission = pd.read_csv(os.path.join(DATA_PATH, 'sample_submission.csv'))\n# submission.toxic = sub\n# submission.toxic.shape","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# submission","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# max(submission.toxic)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#submission.to_csv('submission.csv', index=False)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"history = model.fit(\n    valid_dataset.repeat(),\n    steps_per_epoch=x_valid.shape[0],\n    epochs=3\n    \n)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"submission = pd.read_csv(os.path.join(DATA_PATH, 'sample_submission.csv'))\n\nsub = model.predict(test_dataset, verbose=1)\nsubmission.toxic = sub\nsubmission.toxic.shape","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"submission.to_csv('submission.csv', index=False)","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}