hcrystalball.utils.generate_multiple_tsdata(n_dates=10, n_regions=2, n_plants=3, n_products=4, country='DE', random_state=None)[source]

Provide easy way how to generate dummy data for tests or tutorial purposes

  • n_dates (int) – lenght of one time-series

  • n_regions (int) – number of regions within column ‘Region’

  • n_plants (int) – number of plants within colum ‘Plant’

  • n_products (int) – number of products within colum ‘Product’

  • country (str) – ISO code of country or country-region

  • random_state (int or RandomState instance,) – Control random number generation following the scikit-learn pattern. Default is to use the global Numpy RNG. For reproducible results pass an int. To draw from a given RNG pass a RandomState.


Data with datetime index and following columns - “Region”, “Plant”, “Product” - for partitioning the data - “Country” - for holidays code - “Raining” - as bool exogenous variable - “Quantity” - as target variable

Return type