tf_privacy.synthetic_linearly_separable_data

Generates synthetic train and test data for logistic regression.

tf_privacy.synthetic_linearly_separable_data(
    num_train: int, num_test: int, dimension: int, num_classes: int
) -> Tuple[RegressionDataset, RegressionDataset]

Args
`num_train`	number of training data points.
`num_test`	number of test data points.
`dimension`	the dimension of the classification problem.
`num_classes`	number of classes, assumed to be at least 2.

Returns
`train_dataset`	num_train labeled examples, with unit l2-norm points.
`test_dataset`	num_test labeled examples, with unit l2-norm points.

Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.

Last updated 2024-02-16 UTC.

tf_privacy.synthetic_linearly_separable_data

Args

Returns