Classifies whether a movie review is positive or negative using LSTM Networks.
The project is about classification of reviews of movies as positive or negative. The dataset is taken from the IMDB and has labels as positive or negative. The model is built using LSTM and has got accuracy over 84% with the test data.
- Python: language
- NumPy: library for numerical calculations
- Matplotlib: library for data visualisation
- PIL: Python Image Library for opening and manage different image formats
- Pytorch: a deep learning framework by Facebook AI Research Team for building neural networks
- torchvision: package consists of popular datasets, model architectures, and common image transformations for computer vision
To use this project, clone the repo
git clone https://github.com/Surya-Prakash-Reddy/Movie-Review-Classification.git
After cloning, you can use the Sentiment Cloud.ipynb or Untitled.ipynb notebook to learn or modify. Sentiment Cloud.ipynb is modified from Untitled.ipynb to train on Google Colab.