tf.keras.layers.Average

View source on GitHub

Layer that averages a list of inputs element-wise.

View aliases

Compat aliases for migration

See Migration guide for more details.

tf.compat.v1.keras.layers.Average

tf.keras.layers.Average(
    **kwargs
)

It takes as input a list of tensors, all of the same shape, and returns a single tensor (also of the same shape).

Example:

x1 = np.ones((2, 2))
x2 = np.zeros((2, 2))
y = tf.keras.layers.Average()([x1, x2])
y.numpy().tolist()
[[0.5, 0.5], [0.5, 0.5]]

Usage in a functional model:

input1 = tf.keras.layers.Input(shape=(16,))
x1 = tf.keras.layers.Dense(8, activation='relu')(input1)
input2 = tf.keras.layers.Input(shape=(32,))
x2 = tf.keras.layers.Dense(8, activation='relu')(input2)
avg = tf.keras.layers.Average()([x1, x2])
out = tf.keras.layers.Dense(4)(avg)
model = tf.keras.models.Model(inputs=[input1, input2], outputs=out)

Raises
ValueError	If there is a shape mismatch between the inputs and the shapes cannot be broadcasted to match.

Arguments
**kwargs	standard layer keyword arguments.