tf.keras.metrics.categorical_accuracy

TensorFlow 1 version View source on GitHub

Calculates how often predictions matches one-hot labels.

tf.keras.metrics.categorical_accuracy(
    y_true, y_pred
)

You can provide logits of classes as y_pred, since argmax of logits and probabilities are same.

y_true One-hot ground truth values.
y_pred The prediction values.

Categorical accuracy values.