mlrl.common.cython.output_sampling module

@author: Michael Rapp (michael.rapp.ml@gmail.com)

class mlrl.common.cython.output_sampling.OutputSamplingWithoutReplacementConfig

Bases: object

Allows to configure a method for sampling outputs without replacement.

get_max_samples() → int

Returns the maximum number of outputs that are included in a sample.

Returns:

The maximum number of outputs that are included in a sample

get_min_samples() → int

Returns the minimum number of outputs that are included in a sample.

Returns:

The minimum number of outputs that are included in a sample

get_sample_size() → float

Returns the fraction of outputs that are included in a sample.

Returns:

The fraction of outputs that are included in a sample

set_max_samples(max_samples: int) → OutputSamplingWithoutReplacementConfig

Sets the maximum number of outputs that should be included in a sample.

Parameters:

max_samples – The maximum number of outputs that should be included in a sample. Must be at least get_min_samples() or 0, if the number of outputs should not be restricted

Returns:

An OutputSamplingWithoutReplacementConfig that allows further configuration of the method for sampling outputs

set_min_samples(min_samples: int) → OutputSamplingWithoutReplacementConfig

Sets the minimum number of outputs that should be included in a sample.

Parameters:

min_samples – The minimum number of outputs that should be included in a sample. Must be at least 1

Returns:

An OutputSamplingWithoutReplacementConfig that allows further configuration of the method for sampling outputs

set_sample_size(sample_size: float) → OutputSamplingWithoutReplacementConfig

Sets the fraction of outputs that should be included in a sample.

Parameters:

sample_size – The fraction of outputs that should be included in a sample, e.g., a value of 0.6 corresponds to 60 % of the available outputs. Must be in (0, 1)

Returns:

An OutputSamplingWithoutReplacementConfig that allows further configuration of the method for sampling outputs