Bi-LSTM-MultiHeadAttention-Article-Bias-Prediction.ipynb

{"metadata":{"kaggle":{"accelerator":"gpu","dataSources":[{"sourceId":1246668,"sourceType":"datasetVersion","datasetId":715814},{"sourceId":3670159,"sourceType":"datasetVersion","datasetId":2196935}],"dockerImageVersionId":30665,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":true},"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"name":"python","version":"3.10.13","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"import numpy as np\nimport pandas as pd\nimport torch\nfrom torch import nn\nimport torchmetrics\nfrom tqdm import tqdm\nfrom torchtext.vocab import GloVe, vocab\nfrom torchtext.data.utils import get_tokenizer\nimport matplotlib.pyplot as plt","metadata":{"_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","execution":{"iopub.status.busy":"2024-03-21T14:36:10.682950Z","iopub.execute_input":"2024-03-21T14:36:10.683334Z","iopub.status.idle":"2024-03-21T14:36:10.691461Z","shell.execute_reply.started":"2024-03-21T14:36:10.683300Z","shell.execute_reply":"2024-03-21T14:36:10.690489Z"},"trusted":true},"execution_count":41,"outputs":[]},{"cell_type":"code","source":"train_path = '/kaggle/input/bias-of-us-news-media-houses/Train.xlsx'","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:05:10.503969Z","iopub.execute_input":"2024-03-21T14:05:10.504545Z","iopub.status.idle":"2024-03-21T14:05:10.509260Z","shell.execute_reply.started":"2024-03-21T14:05:10.504509Z","shell.execute_reply":"2024-03-21T14:05:10.508110Z"},"trusted":true},"execution_count":2,"outputs":[]},{"cell_type":"code","source":"df_train = pd.read_excel(train_path,engine='openpyxl')","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:05:10.510445Z","iopub.execute_input":"2024-03-21T14:05:10.510776Z","iopub.status.idle":"2024-03-21T14:05:27.877770Z","shell.execute_reply.started":"2024-03-21T14:05:10.510746Z","shell.execute_reply":"2024-03-21T14:05:27.876952Z"},"trusted":true},"execution_count":3,"outputs":[]},{"cell_type":"code","source":"import re\ndef clean(text):\n    text = re.sub(r\"http\\S+\", \" \", text) # remove urls\n    text = re.sub(r\"RT \", \" \", text) # remove rt\n    text = re.sub(r\"[^a-zA-Z\\'\\.\\,\\d\\s]\", \" \", text) # remove special character except # @ . ,\n    text = re.sub(r'\\t', ' ', text) # remove tabs\n    text = re.sub(r'\\n', ' ', text) # remove line jump\n    text = re.sub(r\"\\s+\", \" \", text) # remove extra white space\n    text = text.strip()\n    return text","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:05:27.880192Z","iopub.execute_input":"2024-03-21T14:05:27.880852Z","iopub.status.idle":"2024-03-21T14:05:27.886792Z","shell.execute_reply.started":"2024-03-21T14:05:27.880814Z","shell.execute_reply":"2024-03-21T14:05:27.885851Z"},"trusted":true},"execution_count":4,"outputs":[]},{"cell_type":"code","source":"df_train['content_original'] = df_train['content_original'].str.split(' \\n\\n---\\n\\n').str[0]\ndf_train['content_original'] = df_train['content_original'].str.replace('-',' ').str.replace('[^\\w\\s]','').str.replace('\\n',' ').str.lower()\n\ndf_train['title'] = df_train['title'].str.split(' \\n\\n---\\n\\n').str[0]\ndf_train['title'] = df_train['title'].str.replace('-',' ').str.replace('[^\\w\\s]','').str.replace('\\n',' ').str.lower()\n\n\ndf_train.content_original = df_train.content_original.apply(clean)\ndf_train.title = df_train.title.apply(clean)","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:05:27.887962Z","iopub.execute_input":"2024-03-21T14:05:27.888230Z","iopub.status.idle":"2024-03-21T14:05:49.401995Z","shell.execute_reply.started":"2024-03-21T14:05:27.888207Z","shell.execute_reply":"2024-03-21T14:05:49.401002Z"},"trusted":true},"execution_count":5,"outputs":[]},{"cell_type":"code","source":"unk_token = \"<unk>\"\nunk_index = 0\nglove_vectors = GloVe(name='6B', dim=100)\nglove_vocab = vocab(glove_vectors.stoi)\nglove_vocab.insert_token(\"<unk>\",unk_index)\n#this is necessary otherwise it will throw runtime error if OOV token is queried \nglove_vocab.set_default_index(unk_index)\npretrained_embeddings = glove_vectors.vectors\npretrained_embeddings = torch.cat((torch.zeros(1,pretrained_embeddings.shape[1]),pretrained_embeddings))","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:05:49.403200Z","iopub.execute_input":"2024-03-21T14:05:49.403514Z","iopub.status.idle":"2024-03-21T14:09:12.063206Z","shell.execute_reply.started":"2024-03-21T14:05:49.403489Z","shell.execute_reply":"2024-03-21T14:09:12.062327Z"},"trusted":true},"execution_count":6,"outputs":[{"name":"stderr","text":".vector_cache/glove.6B.zip: 862MB [02:39, 5.42MB/s]                              \n100%|█████████▉| 399999/400000 [00:27<00:00, 14517.21it/s]\n","output_type":"stream"}]},{"cell_type":"code","source":"pretrained_embeddings[glove_vocab['hello']]","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:09:12.064716Z","iopub.execute_input":"2024-03-21T14:09:12.065045Z","iopub.status.idle":"2024-03-21T14:09:12.114571Z","shell.execute_reply.started":"2024-03-21T14:09:12.065019Z","shell.execute_reply":"2024-03-21T14:09:12.113678Z"},"trusted":true},"execution_count":7,"outputs":[{"execution_count":7,"output_type":"execute_result","data":{"text/plain":"tensor([ 2.0972e-01, -9.7601e-01, -3.4845e-02, -4.4451e-01,  5.7553e-01,\n        -5.5722e-01, -4.8259e-02,  8.8938e-01, -3.8567e-01, -6.0122e-01,\n         1.4689e-01,  2.3241e-01, -2.0206e-01, -1.2475e-01,  2.2172e-01,\n        -7.9488e-01,  1.8551e-04, -4.6308e-01, -7.6648e-02,  1.8384e-01,\n         9.2898e-02,  1.6110e-01,  3.7590e-01, -9.7123e-01,  1.1706e+00,\n        -2.3141e-01, -6.8350e-01, -1.3191e-01, -4.8528e-01,  4.8536e-01,\n         8.7823e-02,  2.4492e-01, -2.8698e-01, -3.9631e-02,  4.8862e-02,\n        -3.0619e-01, -6.2387e-01,  1.9994e-01,  6.4592e-02, -1.0950e-01,\n        -2.6365e-01,  1.3784e-02, -1.1060e+00,  3.0101e-01,  6.1965e-01,\n         1.4128e+00,  5.8870e-01, -2.1358e-01,  5.2177e-01,  7.6162e-01,\n        -2.9929e-01,  4.3983e-01,  1.2749e+00,  7.3642e-01, -9.9978e-01,\n        -1.1642e+00, -2.6165e-01, -5.3110e-01,  3.3104e-03, -4.6568e-01,\n        -3.6465e-01,  4.4717e-02,  7.8665e-03,  6.6572e-01,  2.8863e-01,\n         3.7517e-01, -4.0419e-01, -2.0311e-01,  7.3303e-02, -2.5918e-01,\n        -5.5221e-02,  2.6285e-01,  8.3696e-01, -5.7493e-01, -1.6033e-02,\n        -9.6859e-01,  6.8637e-01, -1.5171e-01,  3.0613e-01, -3.9007e-01,\n         5.7236e-01,  6.2419e-01,  6.6544e-01,  4.6587e-01,  2.7847e-02,\n        -3.0305e-01, -7.6987e-01, -6.2289e-01,  5.8136e-01,  1.1842e+00,\n        -3.4949e-01, -4.6764e-01, -5.3002e-01, -1.0356e-01, -1.5637e-01,\n         5.1217e-01,  2.4016e-01,  3.6450e-01,  9.2802e-01,  2.1022e-01])"},"metadata":{}}]},{"cell_type":"code","source":"pretrained_embeddings.shape","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:09:12.115822Z","iopub.execute_input":"2024-03-21T14:09:12.116481Z","iopub.status.idle":"2024-03-21T14:09:12.122028Z","shell.execute_reply.started":"2024-03-21T14:09:12.116447Z","shell.execute_reply":"2024-03-21T14:09:12.121168Z"},"trusted":true},"execution_count":8,"outputs":[{"execution_count":8,"output_type":"execute_result","data":{"text/plain":"torch.Size([400001, 100])"},"metadata":{}}]},{"cell_type":"code","source":"tokenizer = get_tokenizer(\"basic_english\")\nexample_text = \"hello how are you ?\"\ntokens = tokenizer(example_text)\nindices = glove_vocab(tokens)\nmax_len = 120\npadded_indices = indices[:max_len] + [0] * max(0, max_len - len(indices))\ntext_input = torch.tensor(padded_indices)","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:09:12.123547Z","iopub.execute_input":"2024-03-21T14:09:12.123822Z","iopub.status.idle":"2024-03-21T14:09:12.133019Z","shell.execute_reply.started":"2024-03-21T14:09:12.123800Z","shell.execute_reply":"2024-03-21T14:09:12.132275Z"},"trusted":true},"execution_count":9,"outputs":[]},{"cell_type":"code","source":"text_input.shape","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:09:12.137232Z","iopub.execute_input":"2024-03-21T14:09:12.137542Z","iopub.status.idle":"2024-03-21T14:09:12.151800Z","shell.execute_reply.started":"2024-03-21T14:09:12.137520Z","shell.execute_reply":"2024-03-21T14:09:12.151010Z"},"trusted":true},"execution_count":10,"outputs":[{"execution_count":10,"output_type":"execute_result","data":{"text/plain":"torch.Size([120])"},"metadata":{}}]},{"cell_type":"code","source":"df_train.head(2)","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:09:12.152725Z","iopub.execute_input":"2024-03-21T14:09:12.152951Z","iopub.status.idle":"2024-03-21T14:09:12.179134Z","shell.execute_reply.started":"2024-03-21T14:09:12.152931Z","shell.execute_reply":"2024-03-21T14:09:12.178276Z"},"trusted":true},"execution_count":11,"outputs":[{"execution_count":11,"output_type":"execute_result","data":{"text/plain":"   Unnamed: 0        topic           source  bias  \\\n0           0  immigration  National Review     2   \n1           1      culture   Yahoo! The 360     1   \n\n                                                 url  \\\n0  https://www.nationalreview.com/2018/12/governm...   \n1  https://news.yahoo.com/can-the-developing-worl...   \n\n                                             title        date  \\\n0                          shutdown theater, again  2018-12-12   \n1  can the developing world endure the coronavirus  2020-06-30   \n\n                                             authors  \\\n0  Kevin D. Williamson, Kyle Smith, Andrew C. Mcc...   \n1                                      Mike Bebernes   \n\n                                             content  \\\n0  President Trump and Senate Minority Leader Chu...   \n1  “ The 360 ” shows you diverse perspectives on ...   \n\n                                    content_original              source_url  \\\n0  president trump and senate minority leader chu...  www.nationalreview.com   \n1  the 360 shows you diverse perspectives on the ...      www.news.yahoo.com   \n\n  bias_text                ID  \n0     right  zl7kc7EmAyIdUMIo  \n1    center  xpbjYTJYPdlw6HmJ  ","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>Unnamed: 0</th>\n      <th>topic</th>\n      <th>source</th>\n      <th>bias</th>\n      <th>url</th>\n      <th>title</th>\n      <th>date</th>\n      <th>authors</th>\n      <th>content</th>\n      <th>content_original</th>\n      <th>source_url</th>\n      <th>bias_text</th>\n      <th>ID</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0</td>\n      <td>immigration</td>\n      <td>National Review</td>\n      <td>2</td>\n      <td>https://www.nationalreview.com/2018/12/governm...</td>\n      <td>shutdown theater, again</td>\n      <td>2018-12-12</td>\n      <td>Kevin D. Williamson, Kyle Smith, Andrew C. Mcc...</td>\n      <td>President Trump and Senate Minority Leader Chu...</td>\n      <td>president trump and senate minority leader chu...</td>\n      <td>www.nationalreview.com</td>\n      <td>right</td>\n      <td>zl7kc7EmAyIdUMIo</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>1</td>\n      <td>culture</td>\n      <td>Yahoo! The 360</td>\n      <td>1</td>\n      <td>https://news.yahoo.com/can-the-developing-worl...</td>\n      <td>can the developing world endure the coronavirus</td>\n      <td>2020-06-30</td>\n      <td>Mike Bebernes</td>\n      <td>“ The 360 ” shows you diverse perspectives on ...</td>\n      <td>the 360 shows you diverse perspectives on the ...</td>\n      <td>www.news.yahoo.com</td>\n      <td>center</td>\n      <td>xpbjYTJYPdlw6HmJ</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"cell_type":"code","source":"from collections import Counter","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:09:12.180377Z","iopub.execute_input":"2024-03-21T14:09:12.180984Z","iopub.status.idle":"2024-03-21T14:09:12.184966Z","shell.execute_reply.started":"2024-03-21T14:09:12.180951Z","shell.execute_reply":"2024-03-21T14:09:12.183885Z"},"trusted":true},"execution_count":12,"outputs":[]},{"cell_type":"code","source":"df_train.bias.value_counts()","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:09:12.186058Z","iopub.execute_input":"2024-03-21T14:09:12.186351Z","iopub.status.idle":"2024-03-21T14:09:12.203596Z","shell.execute_reply.started":"2024-03-21T14:09:12.186329Z","shell.execute_reply":"2024-03-21T14:09:12.202730Z"},"trusted":true},"execution_count":13,"outputs":[{"execution_count":13,"output_type":"execute_result","data":{"text/plain":"bias\n2    10241\n0     8861\n1     7488\nName: count, dtype: int64"},"metadata":{}}]},{"cell_type":"code","source":"from sklearn.model_selection import train_test_split","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:09:12.204703Z","iopub.execute_input":"2024-03-21T14:09:12.205024Z","iopub.status.idle":"2024-03-21T14:09:12.952926Z","shell.execute_reply.started":"2024-03-21T14:09:12.204995Z","shell.execute_reply":"2024-03-21T14:09:12.952053Z"},"trusted":true},"execution_count":14,"outputs":[]},{"cell_type":"code","source":"X_train, X_test = train_test_split(df_train, test_size=0.2, random_state=42)","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:09:12.954088Z","iopub.execute_input":"2024-03-21T14:09:12.954719Z","iopub.status.idle":"2024-03-21T14:09:12.979302Z","shell.execute_reply.started":"2024-03-21T14:09:12.954682Z","shell.execute_reply":"2024-03-21T14:09:12.978586Z"},"trusted":true},"execution_count":15,"outputs":[]},{"cell_type":"code","source":"print(X_train.shape)\nprint(X_test.shape)","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:09:12.980619Z","iopub.execute_input":"2024-03-21T14:09:12.980897Z","iopub.status.idle":"2024-03-21T14:09:12.985543Z","shell.execute_reply.started":"2024-03-21T14:09:12.980872Z","shell.execute_reply":"2024-03-21T14:09:12.984564Z"},"trusted":true},"execution_count":16,"outputs":[{"name":"stdout","text":"(21272, 13)\n(5318, 13)\n","output_type":"stream"}]},{"cell_type":"code","source":"from torch.utils.data import Dataset, DataLoader","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:09:12.986627Z","iopub.execute_input":"2024-03-21T14:09:12.986875Z","iopub.status.idle":"2024-03-21T14:09:12.995255Z","shell.execute_reply.started":"2024-03-21T14:09:12.986853Z","shell.execute_reply":"2024-03-21T14:09:12.994481Z"},"trusted":true},"execution_count":17,"outputs":[]},{"cell_type":"code","source":"class NQADataset(Dataset):\n  def __init__(self,data,max_sentence_length):\n\n    self.data = data\n    self.max_sentence_length=max_sentence_length\n    self.tokenizer = get_tokenizer(\"basic_english\")\n    self.vocab = glove_vocab\n   \n\n  def __len__(self):\n    return len(self.data)\n  \n  def __getitem__(self,index : int):\n    data_row = self.data.iloc[index]\n    \n    tokens = self.tokenizer(data_row['content_original'])\n    indices = self.vocab(tokens)\n    padded_indices = indices[:self.max_sentence_length] + [0] * max(0, self.max_sentence_length - len(indices))\n    body = torch.tensor(padded_indices)\n    \n    tokens = self.tokenizer(data_row['title'])\n    indices = self.vocab(tokens)\n    padded_indices = indices[:self.max_sentence_length] + [0] * max(0, self.max_sentence_length - len(indices))\n    title = torch.tensor(padded_indices)\n    \n    labels = torch.tensor(data_row['bias'],dtype=torch.long)\n\n    return dict(body = body, \n                title = title,\n                labels = labels)","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:09:12.996739Z","iopub.execute_input":"2024-03-21T14:09:12.997062Z","iopub.status.idle":"2024-03-21T14:09:13.006481Z","shell.execute_reply.started":"2024-03-21T14:09:12.997032Z","shell.execute_reply":"2024-03-21T14:09:13.005759Z"},"trusted":true},"execution_count":18,"outputs":[]},{"cell_type":"code","source":"import pytorch_lightning as pl","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:09:13.007510Z","iopub.execute_input":"2024-03-21T14:09:13.007804Z","iopub.status.idle":"2024-03-21T14:09:13.357563Z","shell.execute_reply.started":"2024-03-21T14:09:13.007781Z","shell.execute_reply":"2024-03-21T14:09:13.356761Z"},"trusted":true},"execution_count":19,"outputs":[]},{"cell_type":"code","source":"class NQADataModule(pl.LightningDataModule):\n  def __init__(self,train_df,test_df,max_sentence_length,batch_size):\n    super().__init__()\n    \n    self.batch_size = batch_size\n    self.train_df = train_df\n    self.test_df = test_df\n    self.batch_size = batch_size\n    self.max_sentence_length=max_sentence_length\n\n  def setup(self,stage=None):\n    self.train_dataset = NQADataset(self.train_df,self.max_sentence_length)\n    self.test_dataset = NQADataset(self.test_df,self.max_sentence_length)\n\n  def train_dataloader(self):\n    return DataLoader(self.train_dataset,batch_size = self.batch_size,shuffle=True,drop_last=True,num_workers=4)\n\n  def val_dataloader(self):\n    return DataLoader(self.test_dataset,batch_size = self.batch_size,drop_last=True,num_workers=4)\n","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:09:13.358581Z","iopub.execute_input":"2024-03-21T14:09:13.358848Z","iopub.status.idle":"2024-03-21T14:09:13.366014Z","shell.execute_reply.started":"2024-03-21T14:09:13.358825Z","shell.execute_reply":"2024-03-21T14:09:13.365179Z"},"trusted":true},"execution_count":20,"outputs":[]},{"cell_type":"code","source":"BS = 1024","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:09:13.367159Z","iopub.execute_input":"2024-03-21T14:09:13.367428Z","iopub.status.idle":"2024-03-21T14:09:13.378604Z","shell.execute_reply.started":"2024-03-21T14:09:13.367406Z","shell.execute_reply":"2024-03-21T14:09:13.377769Z"},"trusted":true},"execution_count":21,"outputs":[]},{"cell_type":"code","source":"data_module = NQADataModule(X_train,X_test,max_sentence_length=120,batch_size=BS)\ndata_module.setup()","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:09:13.379744Z","iopub.execute_input":"2024-03-21T14:09:13.380000Z","iopub.status.idle":"2024-03-21T14:09:13.388272Z","shell.execute_reply.started":"2024-03-21T14:09:13.379978Z","shell.execute_reply":"2024-03-21T14:09:13.387465Z"},"trusted":true},"execution_count":22,"outputs":[]},{"cell_type":"code","source":"# Iterate through the first batch to see a sample\nfor idx,data in enumerate(data_module.val_dataloader()):\n    print(data['body'].shape)\n    print(data['title'].shape)\n    print(data['labels'].shape)\n    break","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:09:13.389345Z","iopub.execute_input":"2024-03-21T14:09:13.389666Z","iopub.status.idle":"2024-03-21T14:09:16.560134Z","shell.execute_reply.started":"2024-03-21T14:09:13.389632Z","shell.execute_reply":"2024-03-21T14:09:16.558986Z"},"trusted":true},"execution_count":23,"outputs":[{"name":"stdout","text":"torch.Size([1024, 120])\ntorch.Size([1024, 120])\ntorch.Size([1024])\n","output_type":"stream"}]},{"cell_type":"code","source":"data_module.test_dataset[0]","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:09:16.561939Z","iopub.execute_input":"2024-03-21T14:09:16.562270Z","iopub.status.idle":"2024-03-21T14:09:16.574523Z","shell.execute_reply.started":"2024-03-21T14:09:16.562240Z","shell.execute_reply":"2024-03-21T14:09:16.573621Z"},"trusted":true},"execution_count":24,"outputs":[{"execution_count":24,"output_type":"execute_result","data":{"text/plain":"{'body': tensor([   118,      0,   1004,      5,      0,   1326,      3,   5350,      1,\n             38,     14,     52,      7,   2593,      1,     39,    588,     88,\n            424,    401,     46,     36,      0,    742,   1980,     43,   1927,\n              1,     16,    684,  34850,      1,    369,      3,      0,  16375,\n           1887,      1,      7,   8350,    198,   2166,   1887,      3,    193,\n           1822,   1193,      2,     19,     48,      3,      7,   4883,      3,\n            590,   5464,      6,      7,  10468,   2567,     92,      1,   5741,\n           1534,   3723,      3,      0,  60424,     14,      7,    728,      3,\n             26,   1356,      5,    209,  12341,      2,    100,    372,      1,\n             26,   2472,      5,    253,   1587,      4,  13587,    193,   1822,\n             31, 111356,    143,     10,    103,      7,    288,    396,  16375,\n           1887,    896,     55,   4834,      3,    193,  64888,    144,      7,\n          28126,      3,   1338,   4666,      3,     26,    664,    883,      2,\n            155,   1321,     33]),\n 'title': tensor([    0,   482,  2593,    17,     0,   268,     4,  2716,   135, 10468,\n            57,  1534,  1980,     0,     0,     0,     0,     0,     0,     0,\n             0,     0,     0,     0,     0,     0,     0,     0,     0,     0,\n             0,     0,     0,     0,     0,     0,     0,     0,     0,     0,\n             0,     0,     0,     0,     0,     0,     0,     0,     0,     0,\n             0,     0,     0,     0,     0,     0,     0,     0,     0,     0,\n             0,     0,     0,     0,     0,     0,     0,     0,     0,     0,\n             0,     0,     0,     0,     0,     0,     0,     0,     0,     0,\n             0,     0,     0,     0,     0,     0,     0,     0,     0,     0,\n             0,     0,     0,     0,     0,     0,     0,     0,     0,     0,\n             0,     0,     0,     0,     0,     0,     0,     0,     0,     0,\n             0,     0,     0,     0,     0,     0,     0,     0,     0,     0]),\n 'labels': tensor(0)}"},"metadata":{}}]},{"cell_type":"code","source":"pl.seed_everything(123)","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:09:16.575948Z","iopub.execute_input":"2024-03-21T14:09:16.576213Z","iopub.status.idle":"2024-03-21T14:09:16.584831Z","shell.execute_reply.started":"2024-03-21T14:09:16.576188Z","shell.execute_reply":"2024-03-21T14:09:16.584031Z"},"trusted":true},"execution_count":25,"outputs":[{"execution_count":25,"output_type":"execute_result","data":{"text/plain":"123"},"metadata":{}}]},{"cell_type":"code","source":"device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\nprint(device)","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:09:16.585797Z","iopub.execute_input":"2024-03-21T14:09:16.586038Z","iopub.status.idle":"2024-03-21T14:09:16.616216Z","shell.execute_reply.started":"2024-03-21T14:09:16.586017Z","shell.execute_reply":"2024-03-21T14:09:16.615370Z"},"trusted":true},"execution_count":26,"outputs":[{"name":"stdout","text":"cuda\n","output_type":"stream"}]},{"cell_type":"code","source":"class LitLSTM(pl.LightningModule):\n    def __init__(self,num_classes,num_heads,embed_dim,hidd_dim,batch_size):\n        super(LitLSTM, self).__init__()\n        \n        self.hidd_dim = hidd_dim\n        self.n_mha = num_heads\n        self.num_layer = 2\n        self.embed_dim = embed_dim\n        self.batch_size = batch_size\n        \n        self.Embed = nn.Embedding.from_pretrained(pretrained_embeddings, \n                                                  freeze = True, \n                                                  )\n        \n        \n        self.MHA = nn.MultiheadAttention(embed_dim=self.hidd_dim,\n                                         num_heads=self.n_mha,\n                                         batch_first=True)\n\n        \n        self.BiLSTM = nn.LSTM(input_size=self.embed_dim,\n                               hidden_size=self.hidd_dim, \n                               num_layers=self.num_layer, \n                               batch_first=True,\n                               dropout=0.4,\n                               bias = True,\n                               bidirectional=True)\n\n        \n        self.model = nn.Sequential( nn.Dropout(p=0.4),\n                                    nn.ReLU(),\n                                    nn.Linear(self.hidd_dim, 64),\n                                    nn.BatchNorm1d(64),\n                                    nn.Dropout(p=0.4),\n                                    nn.ReLU(),\n                                    nn.Linear(64, 32),\n                                    nn.BatchNorm1d(32),\n                                    nn.Dropout(p=0.4),\n                                    nn.ReLU(),\n                                    nn.Linear(32, num_classes))\n        \n        self.acc = torchmetrics.classification.Accuracy(num_classes=num_classes,task='multiclass')\n        self.trainacc = torchmetrics.classification.Accuracy(num_classes=num_classes,task='multiclass')\n\n        self.lossfn = nn.CrossEntropyLoss(reduction='mean')\n        \n        self.TLOSS = []\n        self.TACC = []\n        self.VLOSS = []\n        self.VACC = []\n        \n        self.validation_step_outputs = []\n        self.train_step_outputs = []\n\n    def forward(self, head, body):\n        \n        head = self.Embed(head)\n        body = self.Embed(body)\n        \n        head, (final_hidden_state_head, final_cell_state_head) = self.BiLSTM(head)\n        body, (final_hidden_state_body, final_cell_state_body) = self.BiLSTM(body)\n        \n        # final_hidden_state_head -> (bidrectional*num_layers,bs,hidden_dim)\n        # after permute -> (bs,bidrectional*num_layers,hidden_dim)\n        \n        \n        h = final_hidden_state_head.permute(1,0,2)\n        b = final_hidden_state_body.permute(1,0,2)\n\n        # nn.MultiheadAttention expects input \n        # in form (seq_length, batch_size, embed_dim)\n        \n        mha,_ = self.MHA(h,b,b)\n        \n        # after taking mean() in second axis -> (bs,hidden_dim)\n\n        \n        mha = mha.mean(dim=1)\n        \n        x = self.model(mha)\n        return x\n\n    def training_step(self, batch, batch_idx):\n        labels = batch['labels']\n        outputs = self.forward(batch['title'], batch['body'])\n        loss = self.lossfn(outputs, labels)\n        self.train_step_outputs.append(loss)\n        self.trainacc.update(outputs, labels)\n        self.log(\"train_loss\", loss)\n        return loss\n    \n    def validation_step(self, batch, batch_idx):\n        labels = batch['labels']\n        outputs = self.forward(batch['title'], batch['body'])\n        loss = self.lossfn(outputs, labels)\n        self.validation_step_outputs.append(loss)\n        self.acc.update(outputs, labels)\n        self.log(\"val_loss\", loss)\n        \n    def on_train_epoch_end(self):\n        mean_val = torch.mean(torch.tensor(self.train_step_outputs))\n        self.train_step_outputs.clear()\n        train_accuracy = self.trainacc.compute()\n        self.TACC.append(train_accuracy)\n        self.TLOSS.append(mean_val)\n        self.trainacc.reset()\n        self.log_dict({\"ETrain_loss\" : mean_val,\"ETrain_Acc\" : train_accuracy},prog_bar=True, on_epoch=True)\n        print(f\"\\n||----------Train Accuracy : {train_accuracy:.4} || Train Loss : {mean_val:.4}-----------||\")\n        print(f\"\\n||--------------------------------------------------------------------------||\")\n\n      \n    def on_validation_epoch_end(self):\n        mean_val = torch.mean(torch.tensor(self.validation_step_outputs))\n        self.validation_step_outputs.clear()\n        val_accuracy = self.acc.compute()\n        self.VACC.append(val_accuracy)\n        self.VLOSS.append(mean_val)\n        self.acc.reset()\n        self.log_dict({\"EVal_loss\" : mean_val,\"EVal_Acc\" : val_accuracy},prog_bar=True,on_epoch=True)\n        print(f\"\\n||----------Val Accuracy : {val_accuracy:.4} || Val Loss : {mean_val:.4}-----------||\")\n    \n           \n    \n    def configure_optimizers(self):\n        return torch.optim.AdamW(self.parameters(),lr=0.001)","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:27:37.812280Z","iopub.execute_input":"2024-03-21T14:27:37.813405Z","iopub.status.idle":"2024-03-21T14:27:37.836496Z","shell.execute_reply.started":"2024-03-21T14:27:37.813367Z","shell.execute_reply":"2024-03-21T14:27:37.835328Z"},"trusted":true},"execution_count":33,"outputs":[]},{"cell_type":"code","source":"model = LitLSTM(num_heads=8,embed_dim=100,hidd_dim=200,num_classes=3,batch_size=BS)","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:27:37.838179Z","iopub.execute_input":"2024-03-21T14:27:37.838483Z","iopub.status.idle":"2024-03-21T14:27:37.876530Z","shell.execute_reply.started":"2024-03-21T14:27:37.838459Z","shell.execute_reply":"2024-03-21T14:27:37.875569Z"},"trusted":true},"execution_count":34,"outputs":[]},{"cell_type":"code","source":"from pytorch_lightning.loggers import TensorBoardLogger\n\nlogger = TensorBoardLogger(\"tb_logs\", name=\"my_model\")\n","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:27:37.877682Z","iopub.execute_input":"2024-03-21T14:27:37.877956Z","iopub.status.idle":"2024-03-21T14:27:37.882648Z","shell.execute_reply.started":"2024-03-21T14:27:37.877933Z","shell.execute_reply":"2024-03-21T14:27:37.881727Z"},"trusted":true},"execution_count":35,"outputs":[]},{"cell_type":"code","source":"trainer = pl.Trainer(devices=-1, \n                     accelerator=\"gpu\",\n                     logger = logger,\n                     max_epochs = 20)","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:27:37.884593Z","iopub.execute_input":"2024-03-21T14:27:37.884831Z","iopub.status.idle":"2024-03-21T14:27:37.943860Z","shell.execute_reply.started":"2024-03-21T14:27:37.884810Z","shell.execute_reply":"2024-03-21T14:27:37.942801Z"},"trusted":true},"execution_count":36,"outputs":[]},{"cell_type":"code","source":"trainer.fit(model,data_module)","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:27:37.945201Z","iopub.execute_input":"2024-03-21T14:27:37.945569Z","iopub.status.idle":"2024-03-21T14:33:07.619252Z","shell.execute_reply.started":"2024-03-21T14:27:37.945537Z","shell.execute_reply":"2024-03-21T14:33:07.618378Z"},"trusted":true},"execution_count":37,"outputs":[{"output_type":"display_data","data":{"text/plain":"Sanity Checking: |          | 0/? [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":""}},"metadata":{}},{"name":"stdout","text":"\n||----------Val Accuracy : 0.3867 || Val Loss : 1.096-----------||\n","output_type":"stream"},{"name":"stderr","text":"/opt/conda/lib/python3.10/site-packages/pytorch_lightning/loops/fit_loop.py:298: The number of training batches (20) is smaller than the logging interval Trainer(log_every_n_steps=50). Set a lower value for log_every_n_steps if you want to see logs for the training epoch.\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Training: |          | 0/? [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"574b5daedfd34cb1a494353754a72d64"}},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"Validation: |          | 0/? [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":""}},"metadata":{}},{"name":"stdout","text":"\n||----------Val Accuracy : 0.3816 || Val Loss : 1.092-----------||\n\n||----------Train Accuracy : 0.4011 || Train Loss : 1.094-----------||\n\n||--------------------------------------------------------------------------||\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Validation: |          | 0/? [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":""}},"metadata":{}},{"name":"stdout","text":"\n||----------Val Accuracy : 0.5318 || Val Loss : 1.017-----------||\n\n||----------Train Accuracy : 0.5126 || Train Loss : 0.9761-----------||\n\n||--------------------------------------------------------------------------||\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Validation: |          | 0/? [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":""}},"metadata":{}},{"name":"stdout","text":"\n||----------Val Accuracy : 0.6021 || Val Loss : 0.8521-----------||\n\n||----------Train Accuracy : 0.5982 || Train Loss : 0.878-----------||\n\n||--------------------------------------------------------------------------||\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Validation: |          | 0/? [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":""}},"metadata":{}},{"name":"stdout","text":"\n||----------Val Accuracy : 0.5867 || Val Loss : 0.8559-----------||\n\n||----------Train Accuracy : 0.6247 || Train Loss : 0.8395-----------||\n\n||--------------------------------------------------------------------------||\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Validation: |          | 0/? [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":""}},"metadata":{}},{"name":"stdout","text":"\n||----------Val Accuracy : 0.6295 || Val Loss : 0.8604-----------||\n\n||----------Train Accuracy : 0.6428 || Train Loss : 0.7993-----------||\n\n||--------------------------------------------------------------------------||\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Validation: |          | 0/? [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":""}},"metadata":{}},{"name":"stdout","text":"\n||----------Val Accuracy : 0.6387 || Val Loss : 0.7697-----------||\n\n||----------Train Accuracy : 0.6517 || Train Loss : 0.777-----------||\n\n||--------------------------------------------------------------------------||\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Validation: |          | 0/? [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":""}},"metadata":{}},{"name":"stdout","text":"\n||----------Val Accuracy : 0.6322 || Val Loss : 0.7871-----------||\n\n||----------Train Accuracy : 0.6598 || Train Loss : 0.7654-----------||\n\n||--------------------------------------------------------------------------||\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Validation: |          | 0/? [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":""}},"metadata":{}},{"name":"stdout","text":"\n||----------Val Accuracy : 0.6596 || Val Loss : 0.7495-----------||\n\n||----------Train Accuracy : 0.6777 || Train Loss : 0.7374-----------||\n\n||--------------------------------------------------------------------------||\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Validation: |          | 0/? [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":""}},"metadata":{}},{"name":"stdout","text":"\n||----------Val Accuracy : 0.6242 || Val Loss : 0.8289-----------||\n\n||----------Train Accuracy : 0.6995 || Train Loss : 0.7027-----------||\n\n||--------------------------------------------------------------------------||\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Validation: |          | 0/? [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":""}},"metadata":{}},{"name":"stdout","text":"\n||----------Val Accuracy : 0.6865 || Val Loss : 0.7117-----------||\n\n||----------Train Accuracy : 0.7116 || Train Loss : 0.6836-----------||\n\n||--------------------------------------------------------------------------||\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Validation: |          | 0/? [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":""}},"metadata":{}},{"name":"stdout","text":"\n||----------Val Accuracy : 0.6496 || Val Loss : 0.8541-----------||\n\n||----------Train Accuracy : 0.7335 || Train Loss : 0.6465-----------||\n\n||--------------------------------------------------------------------------||\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Validation: |          | 0/? [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":""}},"metadata":{}},{"name":"stdout","text":"\n||----------Val Accuracy : 0.6824 || Val Loss : 0.7298-----------||\n\n||----------Train Accuracy : 0.7425 || Train Loss : 0.6286-----------||\n\n||--------------------------------------------------------------------------||\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Validation: |          | 0/? [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":""}},"metadata":{}},{"name":"stdout","text":"\n||----------Val Accuracy : 0.7221 || Val Loss : 0.6679-----------||\n\n||----------Train Accuracy : 0.7509 || Train Loss : 0.6135-----------||\n\n||--------------------------------------------------------------------------||\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Validation: |          | 0/? [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":""}},"metadata":{}},{"name":"stdout","text":"\n||----------Val Accuracy : 0.7098 || Val Loss : 0.6572-----------||\n\n||----------Train Accuracy : 0.7559 || Train Loss : 0.6-----------||\n\n||--------------------------------------------------------------------------||\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Validation: |          | 0/? [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":""}},"metadata":{}},{"name":"stdout","text":"\n||----------Val Accuracy : 0.651 || Val Loss : 0.8934-----------||\n\n||----------Train Accuracy : 0.753 || Train Loss : 0.603-----------||\n\n||--------------------------------------------------------------------------||\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Validation: |          | 0/? [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":""}},"metadata":{}},{"name":"stdout","text":"\n||----------Val Accuracy : 0.7096 || Val Loss : 0.6769-----------||\n\n||----------Train Accuracy : 0.7687 || Train Loss : 0.5751-----------||\n\n||--------------------------------------------------------------------------||\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Validation: |          | 0/? [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":""}},"metadata":{}},{"name":"stdout","text":"\n||----------Val Accuracy : 0.725 || Val Loss : 0.6417-----------||\n\n||----------Train Accuracy : 0.7865 || Train Loss : 0.5388-----------||\n\n||--------------------------------------------------------------------------||\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Validation: |          | 0/? [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":""}},"metadata":{}},{"name":"stdout","text":"\n||----------Val Accuracy : 0.6881 || Val Loss : 0.7859-----------||\n\n||----------Train Accuracy : 0.7842 || Train Loss : 0.5485-----------||\n\n||--------------------------------------------------------------------------||\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Validation: |          | 0/? [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":""}},"metadata":{}},{"name":"stdout","text":"\n||----------Val Accuracy : 0.6984 || Val Loss : 0.6759-----------||\n\n||----------Train Accuracy : 0.7855 || Train Loss : 0.5499-----------||\n\n||--------------------------------------------------------------------------||\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"Validation: |          | 0/? [00:00<?, ?it/s]","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":""}},"metadata":{}},{"name":"stdout","text":"\n||----------Val Accuracy : 0.7359 || Val Loss : 0.6437-----------||\n\n||----------Train Accuracy : 0.8083 || Train Loss : 0.5006-----------||\n\n||--------------------------------------------------------------------------||\n","output_type":"stream"}]},{"cell_type":"code","source":"xx = np.arange(20)\nplt.style.use(\"fivethirtyeight\")\n\nacctr = torch.Tensor(model.TACC).detach().cpu().numpy()\nlsstr = torch.Tensor(model.TLOSS).detach().cpu().numpy()\naccte = torch.Tensor(model.VACC).detach().cpu().numpy()\nlsste = torch.Tensor(model.VLOSS).detach().cpu().numpy()\n\naccte = accte[1:]\nlsste = lsste[1:]\n\nfig, (ax1, ax2) = plt.subplots(1, 2,figsize=(12,6))\nfig.suptitle('ACC vs LOSS')\nax1.plot(xx, acctr,label='Train')\nax1.plot(xx,accte,label='Val')\nax1.legend(loc=\"best\")\nax2.plot(xx, lsstr,label='Train')\nax2.plot(xx, lsste,label='Val')\nax2.legend(loc=\"best\")\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-03-21T14:38:20.381108Z","iopub.execute_input":"2024-03-21T14:38:20.381802Z","iopub.status.idle":"2024-03-21T14:38:20.791296Z","shell.execute_reply.started":"2024-03-21T14:38:20.381768Z","shell.execute_reply":"2024-03-21T14:38:20.790367Z"},"trusted":true},"execution_count":45,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 1200x600 with 2 Axes>","image/png":"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"},"metadata":{}}]}]}