Create bert_multilabel_model.pth

bert_multilabel_model.pth

@@ -0,0 +1,115 @@
+# Install required packages
+!pip install -q transformers scikit-learn
+
+# Imports
+import pandas as pd
+import torch
+import torch.nn as nn
+from torch.utils.data import Dataset, DataLoader
+from sklearn.model_selection import train_test_split
+from sklearn.preprocessing import MultiLabelBinarizer
+from transformers import BertTokenizer, BertModel
+from torch.optim import Adam
+
+# Constants
+TEXT_COLUMN = 'Sanction_Context'
+LABEL_COLUMNS = [
+    'Red_Flag_Reason',
+    'Maker_Action',
+    'Escalation_Level',
+    'Risk_Category',
+    'Risk_Drivers',
+    'Investigation_Outcome'
+]
+
+# Load and clean data
+df = pd.read_csv('/kaggle/input/systhesis/synthetic_transactions_samples_5000.csv')
+df = df.dropna(subset=[TEXT_COLUMN])
+df[LABEL_COLUMNS] = df[LABEL_COLUMNS].fillna('Unknown')  # Fill missing labels
+
+# Encode labels using MultiLabelBinarizer
+mlb = MultiLabelBinarizer()
+Y = mlb.fit_transform(df[LABEL_COLUMNS].astype(str).values.tolist())
+X = df[TEXT_COLUMN].tolist()
+
+# Save the label classes for decoding later
+import pickle
+with open("mlb_classes.pkl", "wb") as f:
+    pickle.dump(mlb.classes_, f)
+
+# Tokenizer
+tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
+
+# Dataset Class
+class BertMultiLabelDataset(Dataset):
+    def __init__(self, texts, labels, tokenizer, max_len=128):
+        self.texts = texts
+        self.labels = labels
+        self.tokenizer = tokenizer
+        self.max_len = max_len
+
+    def __getitem__(self, idx):
+        encoding = self.tokenizer(
+            self.texts[idx],
+            padding='max_length',
+            truncation=True,
+            max_length=self.max_len,
+            return_tensors="pt"
+        )
+        item = {key: val.squeeze(0) for key, val in encoding.items()}
+        item['labels'] = torch.FloatTensor(self.labels[idx])
+        return item
+
+    def __len__(self):
+        return len(self.texts)
+
+# Model Definition
+class BertForMultiLabel(nn.Module):
+    def __init__(self, num_labels):
+        super().__init__()
+        self.bert = BertModel.from_pretrained('bert-base-uncased')
+        self.dropout = nn.Dropout(0.3)
+        self.classifier = nn.Linear(self.bert.config.hidden_size, num_labels)
+
+    def forward(self, input_ids, attention_mask):
+        outputs = self.bert(input_ids=input_ids, attention_mask=attention_mask)
+        pooled_output = self.dropout(outputs.pooler_output)
+        logits = self.classifier(pooled_output)
+        return logits
+
+# Prepare data
+X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.2)
+
+train_dataset = BertMultiLabelDataset(X_train, Y_train, tokenizer)
+train_loader = DataLoader(train_dataset, batch_size=16, shuffle=True)
+
+# Device setup
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+print(f"Training on device: {device}")
+
+# Model, optimizer, loss
+model = BertForMultiLabel(num_labels=Y.shape[1]).to(device)
+optimizer = Adam(model.parameters(), lr=2e-5)
+loss_fn = nn.BCEWithLogitsLoss()
+
+# Training loop
+for epoch in range(3):
+    model.train()
+    total_loss = 0
+    for i, batch in enumerate(train_loader):
+        input_ids = batch['input_ids'].to(device)
+        attention_mask = batch['attention_mask'].to(device)
+        labels = batch['labels'].to(device)
+
+        optimizer.zero_grad()
+        logits = model(input_ids, attention_mask)
+        loss = loss_fn(logits, labels)
+        loss.backward()
+        optimizer.step()
+
+        total_loss += loss.item()
+        if i % 10 == 0:
+            print(f"Epoch {epoch+1}, Step {i}, Loss: {loss.item():.4f}")
+
+    avg_loss = total_loss / len(train_loader)
+    print(f"Epoch {epoch+1} finished. Avg Loss: {avg_loss:.4f}")