delta.data.estimateClassweights

delta.data.estimateClassweights(gene: Iterator, num_samples: int = 30) → Tuple[float, ...]

Estimate the class weights to use with the weighted categorical cross-entropy based on the output of the trainGenerator_track output.

Parameters

genegenerator

Tracking U-Net training generator. (output of trainGenerator_seg/track)

num_samplesint, optional

Number of batches to use for estimation. The default is 30.

Returns

class_weightstuple of floats

Relative weights of each class. Note that, if 0 elements of a certain class are present in the samples, the weight for this class will be set to 0.