https://github.com/martinp95/text_summarization_with_t5forconditionalgeneration

The Text Summarization Project aims to develop an abstractive summarization model using the T5 architecture, fine-tuned on the California state bill subset of the BillSum dataset. The primary goal is to create a tool that can efficiently summarize lengthy texts from newsletters and other sources, providing concise and informative summaries.
https://github.com/martinp95/text_summarization_with_t5forconditionalgeneration

huggingface nlp streamlit summarization t5-model

Last synced: about 1 month ago
JSON representation

The Text Summarization Project aims to develop an abstractive summarization model using the T5 architecture, fine-tuned on the California state bill subset of the BillSum dataset. The primary goal is to create a tool that can efficiently summarize lengthy texts from newsletters and other sources, providing concise and informative summaries.

• Host: GitHub
• URL: https://github.com/martinp95/text_summarization_with_t5forconditionalgeneration
• Owner: martinp95
• License: mit
• Created: 2025-01-09T12:39:38.000Z (over 1 year ago)
• Default Branch: master
• Last Pushed: 2025-01-15T11:20:49.000Z (over 1 year ago)
• Last Synced: 2025-03-15T11:50:00.650Z (over 1 year ago)
• Topics: huggingface, nlp, streamlit, summarization, t5-model
• Language: Python
• Homepage:
• Size: 454 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# Text Summarization with T5ForConditionalGeneration

This project provides a simple summarizer using fine-tuned T5ForConditionalGeneration model to generate text summarice

## Table of Contents

- [About](#about)

- [Installation](#installation)

- [Usage](#usage)

- [License](#license)

## About

The Text Summarization Project aims to develop an abstractive summarization model using the T5 architecture, fine-tuned on the California state bill subset of the BillSum dataset. The primary goal is to create a tool that can efficiently summarize lengthy texts from newsletters and other sources, providing concise and informative summaries.

Key features:

- Upload a text file for summarization.

- Input custom text and generate summaries in real-time.

## Installation

1. Clone the repository:

    ```sh

    git clone https://github.com/martinp95/text_summarization_with_T5ForConditionalGeneration.git

    cd text_summarization_with_T5ForConditionalGeneration

    ```

2. Create and activate the Conda environment:

    ```sh

    conda env create -f environment.yml

    conda activate text_summarization_with_T5ForConditionalGeneration

    ```

## Usage

### Step 1: Fine-Tune the Model

To fine-tune the model, use the Jupyter notebook `train_model.ipynb`. This notebook walks you through the steps of training the T5 model on the desired dataset.

1. Open the notebook

2. Execute the cells sequentially to fine-tune the model.

3. Save the fine-tune model in the `fine_tuned/` direcotry

### Step 2: Run the Application

Once the model is fine-tuned and saved, you can launch the Streamlit application:

1. Start the application:

```sh

cd ./src/

streamlit run app.py

```

2. Open the local URL displayed in the terminal in your browser.

![Summarization APP](/images/text_summarization_application.png)

### Step 3: Using the Application

* Upload a Text File: Upload a `.txt` file containing the text you want to summarize.

![Uploaded File Summarization](/images/upload_file_summarization.png)

* Manual Text Input: Use the input field to type or parte text directly

![Manual Text Summarization](/images/manually_input_summarize.png)

## License

This project is licensed under the MIT License. See the [LICENSE](LICENSE) file for details.