qbiocode.data_generation.make_swiss_roll module#

Summary#

Functions:

my_make_swiss_roll

This function generates a series of 'swiss roll' data sets.

Reference#

my_make_swiss_roll(n_samples=[100, 120, 140, 160, 180, 200, 220, 240, 260, 280], noise=[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9], hole=[True, False], save_path=None)[source]#

This function generates a series of ‘swiss roll’ data sets. It uses itertools to generate a range of input arguments to pass into the sklearn make_swiss_roll function. A data set is generated for each set of input arguments, which allows the function to produce a variety of different swiss rolls based on varying number of samples (n_samples), noise, and whether or not the swiss roll has a ‘hole’ in it.

Parameters:
  • n_samples – list of integers The number of sample points on the Swiss Roll.

  • noise – list of floats The standard deviation of the gaussian noise.

  • hole – list of bools If True generates the swiss roll with hole dataset.

Returns:

Dataset in pandas dataframe with samples in the first column, features in the middle columns, and

labels in the last column.

Return type:

df (pandas.DataFrame)