This project uses a TensorFlow-based model to classify text into positive or negative sentiments. It utilizes LSTM layers for sequence modeling, and the dataset can be any sentiment-labeled text data (e.g., IMDb movie reviews or Twitter data).
data/: Contains raw and processed datasets.src/: Main Python source code including data preprocessing, model definition, training, and evaluation.notebooks/: Jupyter notebooks for EDA.tests/: Unit tests for key components.scripts/: Scripts to run tasks like downloading data and training the model.saved_models/: Directory where the trained models are saved.logs/: Logs generated during model training.
- Clone this repository.
- Install the dependencies:
pip install -r requirements.txt