readme.md

metadata

title: Instamatch
app_file: app.py
sdk: gradio
sdk_version: 5.12.0

Overview

InstaMatch is a cloud instance matching tool for machine learning models. It fetches model information from HuggingFace and recommends suitable cloud instances from AWS, Azure, and GCP based on the model's requirements.

Features

• Fetch model information from HuggingFace API
• Estimate model requirements (vCPUs and memory)
• Recommend suitable cloud instances from AWS, Azure, and GCP
• Display primary and backup recommendations

Installation

1. Clone the repository: ```sh git clone https://github.com/yourusername/InstaMatch.git cd InstaMatch ```

2. Create and activate a virtual environment: ```sh python3 -m venv venv source venv/bin/activate ```

3. Install the required dependencies: ```sh pip install -r requirements.txt ```

4. Set up your HuggingFace API token in a `.env` file: ```sh echo "HUGGING_FACE_TOKEN=your_huggingface_token" > .env ```

Usage

1. Run the application: ```sh python app.py ```

2. Open the provided URL in your browser.

3. Enter a model name from HuggingFace (e.g., `gpt2`, `bert-base-uncased`) and click "Get Recommendations".

4. View the model requirements and cloud instance recommendations.

Files

• `app.py`: Main application file
• `cloud_instances.csv`: CSV file containing cloud instance data
• `requirements.txt`: List of required Python packages
• `utils/`: Utility functions and modules

License

This project is licensed under the MIT License." > README.md

git add README.md git commit -m "Add README file"