Computes the Huber loss between y_true and y_pred.

Used in the notebooks

Used in the tutorials

For each value x in error = y_true - y_pred:

loss = 0.5 * x^2                  if |x| <= d
loss = 0.5 * d^2 + d * (|x| - d)  if |x| > d

where d is delta. See:

Standalone usage:

y_true = [[0, 1], [0, 0]]
y_pred = [[0.6, 0.4], [0.4, 0.6]]
# Using 'auto'/'sum_over_batch_size' reduction type.
h = tf.keras.losses.Huber()
h(y_true, y_pred).numpy()
# Calling with 'sample_weight'.
h(y_true, y_pred, sample_weight=[1, 0]).numpy()
# Using 'sum' reduction type.
h = tf.keras.losses.Huber(
h(y_true, y_pred).numpy()
# Using 'none' reduction type.
h = tf.keras.losses.Huber(
h(y_true, y_pred).numpy()
array([0.18, 0.13], dtype=float32)

Usage with the compile() API:

model.compile(optimizer='sgd', loss=tf.keras.losses.Huber())

delta A float, the point where the Huber loss function changes from a quadratic to linear.
reduction (Optional) Type of tf.keras.losses.Reduction to apply to loss. Default value is AUTO. AUTO indicates that the reduction option will be determined by the usage context. For almost all cases this defaults to SUM_OVER_BATCH_SIZE. When used with tf.distribute.Strategy, outside of built-in training loops such as tf.keras compile and fit, using AUTO or SUM_OVER_BATCH_SIZE will raise an error. Please see this custom training tutorial for more details.
name Optional name for the op. Defaults to 'huber_loss'.



View source

Instantiates a Loss from its config (output of get_config()).

config Output of get_config().

A Loss instance.