tf.keras.utils.to_categorical

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

Used in the notebooks

Used in the guide Used in the tutorials

E.g. for use with categorical_crossentropy.

y 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(y) + 1.
dtype The data type expected by the input. Default: 'float32'.

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

Example:

a = tf.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 = tf.constant([.9, .04, .03, .03,
                 .3, .45, .15, .13,
                 .04, .01, .94, .05,
                 .12, .21, .5, .17],
                shape=[4, 4])
loss = tf.keras.backend.categorical_crossentropy(a, b)
print(np.around(loss, 5))
[0.10536 0.82807 0.1011  1.77196]
loss = tf.keras.backend.categorical_crossentropy(a, a)
print(np.around(loss, 5))
[0. 0. 0. 0.]