SentiToolKit

SentiToolKit  

SentiToolKit is a simple sentiment analysis library that leverages LSTM (Long Short-Term Memory) neural networks for predicting the sentiment of a given text. This toolkit provides a streamlined interface for text preprocessing, model loading, and prediction.

✨ Features

• Pre-trained Model: Uses a trained LSTM model for sentiment analysis.
• Tokenization Support: Handles text tokenization with a customizable vocabulary size.
• Sentiment Prediction: Predicts whether a text is Positive, Neutral, or Negative.
• Easy to Use: Minimal setup required to get predictions.
• Customizable: You can use your own tokenizer and model for fine-tuning.

📦 Installation

You can install the SentiToolKit package via pip:

pip install SentiToolKit==1.0.8

Alternatively, if you'd like to build from source:

• Clone the repository:

  git clone https://github.com/yourusername/SentiToolKit.git
  

• Navigate to the project directory and install the package:

  cd SentiToolKit
  pip install .
  

🚀 Quickstart

To get started with SentiToolKit, follow the example below:

# test_model.py

# Import the SentiToolKit class from your package
from SentiAnalyzer.Main_tensor_model import SentiToolKit

def main():
    # Instantiate the model
    model = SentiToolKit()

    # Get user input for testing
    review = input("Enter a review for sentiment analysis: ")

    # Call the model to get the sentiment
    sentiment = model.__call__(review)

    # Print the result
    print("Sentiment Analysis Result:", sentiment)

if __name__ == "__main__":
    main()

🧰 Usage

• Loading the Pre-trained Model: The toolkit loads a pre-trained LSTM model from the provided .keras file.
• Text Tokenization: The input text is tokenized and padded using the stored tokenizer from tokenizer.pkl.
• Sentiment Prediction: Predicts the sentiment based on the input text:
  • Positive
  • Neutral
  • Negative

📁 Project Structure

SentiToolKit/
├── converter.py                 
├── converter1.py
├── dataExtractor.py             
├── tensor_model.py              
├── usage.py                    
├── json/                        
│   ├── backup.json              
│   ├── negative_reviews.json    
│   ├── neutral_reviews.json     
│   ├── positive_reviews.json    
│   └── train.json               
├── SentiAnalyzer/               
│   ├── __init__.py              
│   ├── Main_tensor_model.py     
│   ├── SentiToolKit.keras       
│   └── tokenizer.pkl            
├── tests/                       
├── text/                        
├── .env                         
├── .gitignore                   
├── LICENSE                      
├── MANIFEST.in                  
├── pyproject.toml               
├── README.md                    
├── SentiToolKit.keras           
├── setup.py                     
└── tokenizer.pkl                

📄 License

This project is licensed under the GNU AFFERO GENERAL PUBLIC LICENSE Version 3,. See the LICENSE file for details.

💡 Future Improvements

• Add support for additional languages.
• Fine-tune the pre-trained model for domain-specific tasks.

👥 Contributing

Contributions are welcome! If you’d like to contribute to SentiToolKit, feel free to fork the repository and submit a pull request.

• Fork the repo.
• Create your feature branch
• Commit your changes
• Push to the branch
• Open a pull request.

📬 Contact

• Author: Niall Dcunha
• GitHub: https://github.com/Niall1985