tarneemalaa/bert_imdb_model

BERT IMDb Sentiment Classifier

Dataset

• IMDb Dataset (50,000 reviews: 25k train + 25k test)
• Source: IMDB Dataset of 50K Movie Reviews Kaggle
• Preprocessing: Tokenization via BERT tokenizer

Model Architecture

• Base: bert-base-uncased from Hugging Face Transformers
• Fine-tuned with a classification head (linear layer)
• Trained with PyTorch using AdamW optimizer and scheduler