Experiments

Description of Train and Test Data

Todo

  • Simple test-train split.
  • Train data from 2017.
  • Test data from 2018.
  • Calendar heatmap graphic.

Evaluation Metrics

Todo

  • MAE vs. MSE
  • MAPE vs. sMAPE
  • R^2
  • day-night vs day-only

Preprocess & Feature Selection

Todo

  • What features are useful?
  • Computed time-of-day and time-of-year features.
  • Training with day-night vs day-only.

Empirical Learning Curve

Todo

  • Error vs amount of training data used.
  • Linear regression vs. Random Forest vs. GBTs.

Reference Time vs Forecast Time

Todo

  • Error for each (reftime, forecast time) pair.
  • Heatmap.