How to add a dataset

Dataset Format

All datasets are stored in the data/ folder in .csv format. Each file is named using the pattern [dataset_id]_[dataset_name].csv, e.g., ds4_ashd_with_weather.csv.

Some datasets may share the same net load data but differ in the availability of exogenous variables. For instance, ds1_ashd.csv is equivalent to ds4_ashd_with_weather.csv but without weather data.

Each CSV file must include two required columns: datetime and netload_kW.
PyNNLF uses netload_kW as the target variable for forecasting, and automatically generates lag features based on it.

Any additional columns are treated as exogenous variables. These are also processed into lag features based on the forecast horizon, but are not used as targets.

Calendar features are excluded from the CSV files, as PyNNLF generates them dynamically during each experiment.

How to Add a Dataset

To add a new dataset, simply create a .csv file in the data/ folder following the naming convention above.
Make sure to update data/metadata.xlsx to document the new dataset.