Toxic-comment-classification

Description

This project aims to classify toxic comments using machine learning techniques. It provides a solution to identify and filter out toxic comments from data collected on Kaggle competition Toxic Comment Classification Challenge.

Features

• Preprocessing of text data to remove noise and irrelevant information.
• Training and evaluation of machine learning models for toxic comment classification.
• Integration with a user interface for easy interaction and input of comments.

Installation

1. Clone the repository:

   git clone https://github.com/your-username/Toxic-comment-classification.git
   cd Toxic-comment-classification

2. Install the required dependencies:

   pip install -r requirements.txt

3. If you have a capable GPU, use a conda environment is advised:

   conda env create -f environment.yaml

   or if you prefer pip:

   pip install -r requirements2.txt

Usage

1. Navigate to the project directory:

   cd Toxic-comment-classification

2. Inference:

   python demo/Demo_GUI.py

3. To try the demo, download the models folder from this link and put it in the root folder. The model_checkpoint folder should be directly under the Toxic-comment-classification folder.

4. To train the models, download the .vec and .txt embedding files from the internet, and put it in the folder like this:

Contributing

Contributions are welcome! If you find any issues or have suggestions for improvements, please open an issue or submit a pull request.

License

This project is licensed under the MIT License. See the LICENSE file for more details.