README.md

metadata

title: Bayesian Token Co-occurrence Simulator
emoji: 🧮
colorFrom: blue
colorTo: indigo
sdk: docker
sdk_version: 3.0.0
app_file: app.py
pinned: false

Bayesian Token Co-occurrence Simulator

A Streamlit application that simulates and visualizes token co-occurrence patterns using Bayesian methods.

Local Development Setup

1. Create and activate a virtual environment:

# Create virtual environment
python -m venv venv

# Activate virtual environment
# On macOS/Linux:
source venv/bin/activate
# On Windows:
.\venv\Scripts\activate

2. Install dependencies:

pip install -r requirements.txt

3. Run the application:

streamlit run app.py

The application will be available at http://localhost:8501

Docker Setup

1. Build the Docker image:

docker build -t bayesian-token-simulator .

2. Run the Docker container:

docker run -p 8501:8501 bayesian-token-simulator

The application will be available at http://localhost:8501

Hugging Face Spaces Deployment

1. Clone your Hugging Face Space repository:

git clone https://huggingface.co/spaces/afscomercial/stf_model

2. Copy your project files to the cloned repository:

cp -r app.py requirements.txt Dockerfile .huggingfaceignore stf_model/

3. Push the changes to Hugging Face:

cd stf_model
git add .
git commit -m "Add Bayesian Token Co-occurrence Simulator"
git push

Your application will be automatically deployed to https://huggingface.co/spaces/afscomercial/stf_model

Project Structure

• app.py: Main Streamlit application
• requirements.txt: Python dependencies
• Dockerfile: Docker configuration
• .gitignore: Git ignore rules
• .huggingfaceignore: Hugging Face ignore rules

Features

• Interactive text input for training sentences
• Bayesian smoothing parameter adjustment
• Co-occurrence matrix visualization
• Next token prediction
• Real-time updates

Requirements

• Python 3.8+
• Streamlit
• NumPy
• Pandas
• Matplotlib
• Seaborn
• NLTK