Missed TensorFlow Dev Summit? Check out the video playlist. Watch recordings

tf.keras.losses.categorical_crossentropy

TensorFlow 1 version View source on GitHub

Computes the categorical crossentropy loss.

tf.keras.losses.categorical_crossentropy(
    y_true, y_pred, from_logits=False, label_smoothing=0
)

Args:

  • y_true: tensor of true targets.
  • y_pred: tensor of predicted targets.
  • from_logits: Whether y_pred is expected to be a logits tensor. By default, we assume that y_pred encodes a probability distribution.
  • label_smoothing: Float in [0, 1]. If > 0 then smooth the labels.

Returns:

Categorical crossentropy loss value.