TextCondenser is a Python-based text summarization engine that uses Transformer models to distill key information from extensive text sources.
- Ensure you have Python 3.x installed on your system.
- Install the required dependencies using the following command:
pip install -r requirements.txt
- Place your input text in the
data/input_text.txt
file. - Run the
main.py
script to generate a summary:python main.py
- The generated summary will be saved in
data/summaries/summary.txt
.
transformers>=4.11.3
: For Transformer model implementation.torch>=1.10.0
: PyTorch deep learning library.numpy>=1.21.2
: Fundamental package for numerical computing.