gemmr.data.generate_example_dataset

gemmr.data.generate_example_dataset(model, px=5, py=5, ax=0, ay=0, r_between=0.3, n=1000, random_state=0)

Convenience function returning an example dataset for use with CCA or PLS.

Parameters:
  • model ("cca" or "pls") – model for which example data is returned
  • px (int) – number of features in dataset X
  • py (int) – number of features in dataset Y
  • ax (float < 0) – prinicpal component spectrum decay constant for X
  • ay (float < 0) – prinicpal component spectrum decay constant for Y
  • r_between (float between 0 and 1) – assumed true correlation between weighted composites of X and Y
  • n (int) – number of samples to be returned
  • random_state (None, int or random-number-generator instance) – for random number generator initialization
Returns:

  • X (np.ndarray (n_samples, n_features)) – dataset X
  • Y (np.ndarray (n_samples, n_features)) – dataset Y