tf.keras.utils.to_categorical | TensorFlow v2.16.1 (original) (raw)

tf.keras.utils.to_categorical

Converts a class vector (integers) to binary class matrix.

tf.keras.utils.to_categorical(
    x, num_classes=None
)

Used in the notebooks

Used in the tutorials
Implement Differential Privacy with TensorFlow Privacy Assess privacy risks with the TensorFlow Privacy Report On-Device Training with TensorFlow Lite Human Pose Classification with MoveNet and TensorFlow Lite Classifying CIFAR-10 with XLA

E.g. for use with categorical_crossentropy.

Args
x	Array-like with class values to be converted into a matrix (integers from 0 to num_classes - 1).
num_classes	Total number of classes. If None, this would be inferred as max(x) + 1. Defaults to None.

Returns
A binary matrix representation of the input as a NumPy array. The class axis is placed last.

Example:

a = keras.utils.to_categorical([0, 1, 2, 3], num_classes=4) print(a) [[1. 0. 0. 0.] [0. 1. 0. 0.] [0. 0. 1. 0.] [0. 0. 0. 1.]]

b = np.array([.9, .04, .03, .03, .3, .45, .15, .13, .04, .01, .94, .05, .12, .21, .5, .17], shape=[4, 4]) loss = keras.ops.categorical_crossentropy(a, b) print(np.around(loss, 5)) [0.10536 0.82807 0.1011 1.77196]

loss = keras.ops.categorical_crossentropy(a, a) print(np.around(loss, 5)) [0. 0. 0. 0.]

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. Some content is licensed under the numpy license.

Last updated 2024-06-07 UTC.