mltk.core.MltkDataset¶

class MltkDataset[source]¶

Helper class for configuring a training dataset

Refer to mltk.core.DatasetMixin for more details.

Methods

`__init__`
`load_dataset`	Load the dataset subset
`summarize_class_counts`	Generate a text summary of the given class counts dictionary
`summarize_dataset`	Return a string summary of the dataset
`unload_dataset`	Unload the dataset

load_dataset(subset, test=False, **kwargs)[source]¶

Load the dataset subset

This is called automatically by the MLTK before training or evaluation.

Parameters:

subset (str) – The dataset subset to return: ‘training’ or ‘evaluation’
test (bool) – This is optional, it is used when invoking a training “dryrun”, e.g.: mltk train basic_example-test If this is true, then only return a small portion of the dataset for testing purposes

Returns:

None, in this case the function is expected to set the my_model.x, my_model.y, and/or my_model.validation_data properties
x - The x samples, in this case the my_model.x properties will be automatically set
(x, y) - The x samples and y labels. In this case the my_model.x and my_model.y properties will be automatically set
(x, None, validation_data) - The x samples and validation_data. In this case the my_model.x and my_model.validation_data properties will be automatically set

Return type:

One of the following

summarize_dataset()[source]¶

Return a string summary of the dataset

static summarize_class_counts(class_counts)[source]¶

Generate a text summary of the given class counts dictionary