Text Classification using Mamba with IMDB Dataset

Overview

This project aims to perform sentiment analysis on the IMDB movie review dataset using the Mamba Model. The goal is to classify movie reviews as either positive or negative based on their textual content.

Dataset

The IMDB dataset consists of 50,000 movie reviews, split evenly into 25k for training and 25k for testing. Each review is labeled as either positive or negative.

Installation

To run the project locally, follow these steps:

1. Clone this repository:

git clone https://github.com/VuBacktracking/mamba-text-classification.git

2. Install the required dependencies:

pip install -r requirements.txt

Usage

1. Navigate to the project directory:

cd mamba-text-classification
python trainer.py

History of my training

Step	Training Loss	Validation Loss	Accuracy
625	0.020500	0.246246	0.928000
1250	0.671000	0.195849	0.940800
1875	0.596100	0.266093	0.934400
2500	0.016700	0.217099	0.941200
3125	0.000700	0.209536	0.944800
3750	2.680700	0.188751	0.949200
4375	0.015500	0.224948	0.950000
5000	0.002100	0.199092	0.952800
5625	0.013400	0.192042	0.952400
6250	0.152500	0.190083	0.953600

Note: You can check my model on hugging face hub in the link: https://huggingface.co/vubacktracking/mamba_text_classification

Dependencies

• Python 3.x
• Other dependencies listed in requirements.txt

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

• The IMDB dataset: http://ai.stanford.edu/~amaas/data/sentiment/
• Mamba: Mamba: Linear-Time Sequence Modeling with Selective State Spaces