{"cells":[{"metadata":{},"cell_type":"markdown","source":"# Checking the directory and imported files"},{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"cell_type":"code","source":"import os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        print(os.path.join(dirname, filename))","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Importing necessary libraries and packages"},{"metadata":{"_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","trusted":true},"cell_type":"code","source":"import numpy as np\nimport pandas as pd\nimport tensorflow as tf\nimport transformers\nfrom tensorflow.keras.layers import Dense, Input\nfrom tensorflow.keras.optimizers import Adam\nfrom tensorflow.keras.models import Model\nfrom tensorflow.keras.callbacks import ModelCheckpoint\nfrom kaggle_datasets import KaggleDatasets\nfrom transformers import TFAutoModel, AutoTokenizer\nfrom tqdm.notebook import tqdm\nfrom tokenizers import Tokenizer, models, pre_tokenizers\nfrom tokenizers import decoders, processors","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Importing train,test and validation datasets"},{"metadata":{"trusted":true},"cell_type":"code","source":"train1 = pd.read_csv(\"/kaggle/input/jigsaw-multilingual-toxic-comment-classification/jigsaw-toxic-comment-train.csv\")\ntrain2 = pd.read_csv(\"/kaggle/input/jigsaw-multilingual-toxic-comment-classification/jigsaw-unintended-bias-train.csv\")\ntrain2['toxic'] = train2.toxic.round().astype(int)\n\nvalidation_data = pd.read_csv('/kaggle/input/jigsaw-multilingual-toxic-comment-classification/validation.csv')\ntest_data = pd.read_csv('/kaggle/input/jigsaw-multilingual-toxic-comment-classification/test.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":"train2.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"validation_data.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"validation_data['lang'].value_counts()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Let's use only the comments and toxic column from train1 and train2."},{"metadata":{"trusted":true},"cell_type":"code","source":"train = pd.concat([train1[['comment_text', 'toxic']],train2[['comment_text', 'toxic']].query('toxic==1'),train2[['comment_text', 'toxic']].query('toxic==0').sample(n=100000, random_state=0)])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train.toxic.value_counts()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Toggling the TPU Button literally does not turn ON them! Run the below lines to initialize the TPU (Thanks to Google!!)"},{"metadata":{"trusted":true},"cell_type":"code","source":"try:\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    strategy = tf.distribute.get_strategy()\n\nprint(\"REPLICAS: \", strategy.num_replicas_in_sync)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def func1(texts, tokenizer,chunk_size=256,max_len=512):\n    tokenizer.enable_truncation(max_length=max_len)\n    tokenizer.enable_padding(max_length=max_len)\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_func(transformer, max_len=512):\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    history = Dense(1, activation='sigmoid')(cls_token)\n    \n    model = Model(inputs=input_word_ids, outputs=history)\n    model.compile(Adam(lr=1e-4), loss='binary_crossentropy', metrics=['accuracy'])\n    \n    return model","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"AUTO = tf.data.experimental.AUTOTUNE\nKDP = KaggleDatasets().get_gcs_path()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# The model we are going to use is XML_RoBERTa! Thanks to Facebook AI "},{"metadata":{"trusted":true},"cell_type":"code","source":"MODEL = 'jplu/tf-xlm-roberta-large'\ntokenizer = AutoTokenizer.from_pretrained(MODEL)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def func2(texts, tokenizer, maxlen=512):\n    enc_di = tokenizer.batch_encode_plus(\n        texts, \n        return_attention_masks=False, \n        return_token_type_ids=False,\n        pad_to_max_length=True,\n        max_length=maxlen\n    )\n    \n    return np.array(enc_di['input_ids'])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"x_train = func2(train['comment_text'].values, tokenizer, maxlen=192)\nx_valid = func2(validation_data['comment_text'].values, tokenizer, maxlen=192)\nx_test = func2(test_data['content'].values, tokenizer, maxlen=192)\n\ny_train = train['toxic'].values\ny_valid = validation_data['toxic'].values","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"BATCH_SIZE = 16 * strategy.num_replicas_in_sync\nbs = BATCH_SIZE","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_data = (\n    tf.data.Dataset\n    .from_tensor_slices((x_train, y_train))\n    .repeat()\n    .shuffle(2048)\n    .batch(bs)\n    .prefetch(AUTO)\n)\n\nvalid_data = (\n    tf.data.Dataset\n    .from_tensor_slices((x_valid, y_valid))\n    .batch(bs)\n    .cache()\n    .prefetch(AUTO)\n)\n\ntest_data = (\n    tf.data.Dataset\n    .from_tensor_slices(x_test)\n    .batch(bs)\n)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"with strategy.scope():\n    transformer_layer = TFAutoModel.from_pretrained(MODEL)\n    model = build_model_func(transformer_layer, max_len=192)\nmodel.summary()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"steps = x_train.shape[0] // bs\nhistory = model.fit(\n    train_data,\n    steps_per_epoch=steps,\n    validation_data=valid_data,\n    epochs= 1\n)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"steps = x_valid.shape[0] // bs\nhistory_2 = model.fit(\n    valid_data.repeat(),\n    steps_per_epoch=steps,\n    epochs=10\n)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sample_sub=pd.read_csv('../input/jigsaw-multilingual-toxic-comment-classification/sample_submission.csv')\nsample_sub.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"sample_sub['toxic'] = model.predict(test_data, verbose=1)\nsample_sub.to_csv('sample_submission.csv', index=False)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"Sincere Thanks to Mr.Xhulu, whose notebooks have helped me infer knowledge of RoBERTa and its types!\nhttps://www.kaggle.com/xhlulu/jigsaw-tpu-distilbert-with-huggingface-and-keras and \nhttps://www.kaggle.com/xhlulu/jigsaw-tpu-xlm-roberta"},{"metadata":{},"cell_type":"markdown","source":"**Do upvote my notebook to encourage and support me in my early Kaggle days!**"}],"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}