Winter GMU45 Dashboard Project

Introduction

After recovering from a running injury, I set a new goal: to complete the Winter Green Man Ultra (https://www.greenmanultra.co.uk/the-winter-green-man-ultra/) in March 2025. Combining my passion for running and data science, I've embarked on a project to build a personalized training dashboard using data from my Garmin watch.

This project aims to:

• Optimize Training: Provide a customized overview of my running data to enhance my training regimen.
• Demonstrate Data Skills: Apply and showcase my data analysis and visualization expertise through a real-world application.
• Create a Versatile Tool: Develop a dashboard that could potentially benefit other Garmin users seeking personalized data insights.
• More to come...

Table of Contents

• Introduction
• Features
• Technologies Used
• Installation
• Usage
• Project Progress
  • Completed Milestones
  • Data Insights
  • Next Steps
• Contributing
• License
• Contact

Features

• Automated Data Retrieval: Batch scraping of Garmin activity data using an extended functionality of the garmiconnect package.
• Data Processing and Storage: Parsing and storing data from various file formats (.FIT, .TCX, .GPX, .CSV).
• Interactive Dashboard: Built with Dash and Plotly, featuring:
  • Weekly Mileage Tracker: Visualizes weekly running mileage against a 10% incremental target leading up to race day.
  • Countdown Timer: Displays the time remaining until the Winter Green Man Ultra 2025.
  • Route Mapping: Visualizes running routes using GeoPandas and GPX data.
• Scalable Architecture: Plans to implement AWS Lambda functions for automated data updates.

Technologies Used

• Programming Languages: Python
• Data Processing: Pandas, GeoPandas
• Data Visualization: Plotly, Dash
• Web Framework: Dash (by Plotly)
• Data Retrieval: garminconnect package
• Cloud Services (Planned): AWS Lambda
• File Formats: .FIT, .TCX, .GPX, .CSV

Installation

To set up the project locally:

1. Clone the Repository:

   git clone https://github.com/yourusername/winter-gmu45-dashboard.git
   
   

Project Progress

Progress So Far

Project Planning

• Brainstormed End-Product:
  • Envisioned a deployed web app featuring a dashboard with my data, updating daily or semi-daily with the latest Garmin data.

Data Extraction

• Utilized garminexport Package:
  • Found and extended the existing GitHub package garminexport to allow batch scraping.
• Data Retrieval:
  • Successfully pulled all of my activity data from Garmin, obtaining .csv, .tcx, .gpx, and .zip files.

Data Exploration

• File Content Analysis:
  • Explored the data within the files to determine relevance for the dashboard.
• Route Plotting:
  • Experimented with plotting GPX files using GeoPandas for route visualization.

Dashboard Development

• Template Creation:
  • Developed a template Dash application to work towards the end product.
• Weekly Mileage Tracker:
  • Generated the first useful graph—a weekly mileage tracker versus a 10% increase limit until the race date—using Pandas and Plotly Go.
• Dashboard Integration:
  • Placed the graph on the dashboard along with a countdown timer until the race.

---

Data Content Findings

• .FIT Files:
  • Provide the most detailed and comprehensive data, including advanced fitness metrics.
  • Useful for adding heart rate, cadence, power, and other advanced metrics to the dashboard.
• .TCX Files:
  • Similar data to .FIT files but in a more readable XML format.
  • Less efficient for large datasets but easier to inspect manually.
• .GPX Files:
  • Focus on GPS location and route data.
  • Useful for visualizing routes but lack detailed fitness metrics.
• .CSV Files:
  • Good for summary statistics.
  • Lack detailed data such as heart rate or cadence.

---

Next Steps

Automation

• AWS Lambda Functions:
  • Learn how to use AWS Lambda functions to automate data retrieval through the Garmin API.

Dashboard Enhancement

• Visual Improvements:
  • Make the dashboard more visually appealing and improve the user interface.
• Additional Features:
  • Add more features to the dashboard, such as:
    • Heart rate zone analysis.
    • Elevation gain over time.
    • Interactive route maps.

Scalability

• Customization for Other Users:
  • Once the dashboard is polished, useful, and has robust data retrieval, consider making it customizable for other Garmin users.

Advanced Analytics

• Predictive Analytics:
  • Incorporate predictive analytics to forecast performance.
• Metric Correlation Analysis:
  • Analyze correlations between different training metrics.