Running pace prediction from past training patterns, environmental factors, and heart rate

(A) Data Collection
(B) Data Preprocessing
(C) Data analysis
(D) Data Visualization
weather data
fitness data
Japan Meteorological Agency
data scraping
data parsing
library - requests
library - bs4
library - datetime
library - pytz
library - pandas
Author

Mai Tanaka

Abstract
Current progress on the analysis of my running data from training for two marathons. Modeling my running pace per run using recent training data, temperature, relative humidity, and heart rate during the run.
Warning

This project is still ongoing. Documentation is incomplete.

So, What Am I Up To?

  • Developing Python workflows to collect, clean, and integrate Fitbit’s physiological data with meteorological observations from the Japan Meteorological Agency.
  • Exploratory analysis of heart rate, environmental conditions, and training progression to investigate its association with long distance running pace.
  • Building predictive models to examine relationships among physiological signals, environmental stressors, and running pace.
  • Generating visuals to illustrate the insights gained from the analysis
    • Spoiler alert! Preliminary findings suggest that the key to predicting my running pace is to “Train light. Run hard.”