Garmin Running LLM Analysis 🏃

By: Braveenth Rasanayagam

---

The goal of this Jupyter Notebook is to have reuseable code that facilitates the prompting and analysis of your your Garmin-tracked runs by LLMs.

This was built to help train for a half-marathon, and you can use it for your fitness activities.

I solved the context length limitations imposed by various LLM models by splitting the data using even rows, which is a method to reduce the number of data points. I am also considering other mathematical methods of reducing the number of data points.

Please note that I am not responsible or liable for your use of this application. Please use this at your own risk and consult with a health care professional.

Steps to Use:

Using OpenAI's ChatGPT API:

1. Obtain an OpenAI API Key here: https://platform.openai.com/api-keys
2. Add OpenAI API Key as a Colab secret value named 'OPENAI_API_KEY'
3. Obtain .fit file from Garmin Connect
4. Upload .fit file to Colab files
5. Specify path to .fit file in notebook form

Steps to Contribute:

Submit a pull-request or message me! MIT License

Additional Links

• runw.ai
• Instagram - runw.ai