Update README.md

README.md

@@ -1,10 +1,118 @@
----
-title: README
-emoji: 🏃
-colorFrom: purple
-colorTo: gray
-sdk: static
-pinned: false
----
-
-Edit this `README.md` markdown file to author your organization card.
+# -*- coding: utf-8 -*-
+"""CIS 5190 Transformer Model
+
+Automatically generated by Colab.
+
+Original file is located at
+    https://colab.research.google.com/drive/1Iy-nQcufaF7--hI9He7Dp9FsW1TomgrP
+"""
+
+import pandas as pd
+from sklearn.model_selection import train_test_split
+from sklearn.metrics import accuracy_score, classification_report
+import torch
+from torch.utils.data import Dataset, DataLoader
+from transformers import BertTokenizer, BertForSequenceClassification, AdamW
+from transformers import get_scheduler
+from google.colab import drive
+
+drive.mount('/content/drive')
+
+# SET DATASET PATH HERE
+dataset_path = '/content/drive/My Drive/24 Fall/CIS 5190/CIS 5190 Final Project/test_data_random_subset.csv'
+
+news_df = pd.read_csv(dataset_path)
+
+X = news_df['title']
+y = news_df['labels']
+
+# y = y.apply(lambda x: 1 if x == 'FoxNews' else 0)
+
+# Split the data into training and testing sets (80% train, 20% test)
+X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42, shuffle=True, stratify=y)
+
+# Tokenize the text using a BERT tokenizer
+tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
+
+def tokenize_data(texts, tokenizer, max_len=128):
+    return tokenizer(
+        list(texts),
+        padding=True,
+        truncation=True,
+        max_length=max_len,
+        return_tensors="pt"
+    )
+
+# Tokenize the training and test datasets
+train_encodings = tokenize_data(X_train, tokenizer)
+test_encodings = tokenize_data(X_test, tokenizer)
+
+# Create a custom dataset class
+class NewsDataset(Dataset):
+    def __init__(self, encodings, labels):
+        self.encodings = encodings
+        self.labels = labels
+
+    def __len__(self):
+        return len(self.labels)
+
+    def __getitem__(self, idx):
+        item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
+        item['labels'] = torch.tensor(self.labels[idx])
+        return item
+
+train_dataset = NewsDataset(train_encodings, y_train.tolist())
+test_dataset = NewsDataset(test_encodings, y_test.tolist())
+
+# Load DataLoader for batching
+train_loader = DataLoader(train_dataset, batch_size=16, shuffle=True)
+test_loader = DataLoader(test_dataset, batch_size=16)
+
+# Define the model
+model = BertForSequenceClassification.from_pretrained('bert-base-uncased', num_labels=2)
+
+device = torch.device('cuda') if torch.cuda.is_available() else torch.device('cpu')
+model.to(device)
+
+# Define optimizer and scheduler
+optimizer = AdamW(model.parameters(), lr=5e-5)
+num_training_steps = len(train_loader) * 4  # Assume 4 epochs
+lr_scheduler = get_scheduler("linear", optimizer=optimizer, num_warmup_steps=0, num_training_steps=num_training_steps)
+
+# Train the model
+def train_model(model, train_loader, optimizer, scheduler, epochs=4):
+    model.train()
+    for epoch in range(epochs):
+        epoch_loss = 0
+        for batch in train_loader:
+            batch = {k: v.to(device) for k, v in batch.items()}
+            outputs = model(**batch)
+            loss = outputs.loss
+            loss.backward()
+            optimizer.step()
+            scheduler.step()
+            optimizer.zero_grad()
+            epoch_loss += loss.item()
+        print(f"Epoch {epoch+1}/{epochs}, Loss: {epoch_loss/len(train_loader):.4f}")
+
+# Evaluate the model
+def evaluate_model(model, test_loader):
+    model.eval()
+    y_true, y_pred = [], []
+    with torch.no_grad():
+        for batch in test_loader:
+            batch = {k: v.to(device) for k, v in batch.items()}
+            outputs = model(**batch)
+            logits = outputs.logits
+            predictions = torch.argmax(logits, dim=-1)
+            y_true.extend(batch['labels'].tolist())
+            y_pred.extend(predictions.tolist())
+    return y_true, y_pred
+
+train_model(model, train_loader, optimizer, lr_scheduler)
+
+y_true, y_pred = evaluate_model(model, test_loader)
+
+# 11. Print evaluation metrics
+print(f"Accuracy: {accuracy_score(y_true, y_pred):.4f}")
+print("Classification Report:\n", classification_report(y_true, y_pred))