TextClassRNN

TextClassRNN

A text classification project implementing various neural network architectures (RNN, LSTM, FFNN) for sentiment analysis on review data. The models are built using PyTorch and can classify reviews into 5 star ratings (1-5).

Models Implemented

• RNN (Recurrent Neural Network): Basic RNN implementation with tanh activation
• LSTM (Long Short-Term Memory): Advanced RNN architecture with better gradient flow
• FFNN (Feed-Forward Neural Network): Simple baseline model

Requirements

All dependencies are listed in requirements.txt and can be installed using:

pip install -r requirements.txt

Key Features

• Multiple neural network architectures for comparison
• GPU support for faster training
• Word embeddings using GloVe
• Comprehensive evaluation metrics
• Data visualization tools for analysis

Project Structure

• rnn.py: Main RNN implementation
• lstm.py: LSTM model implementation
• ffnn.py: Feed-forward neural network implementation
• eda.py: Exploratory data analysis and visualization
• analysis.py: Training results analysis
• rnn_gpu.py: GPU-optimized RNN implementation

Usage

python rnn.py --train_data path/to/training.json --val_data path/to/validation.json --hidden_dim 20 --epochs 30

Results

The models are evaluated on both training and validation sets, with metrics including:

• Loss
• Training accuracy
• Validation accuracy

Results are saved to CSV files for further analysis and visualization.