tf.keras.activations.tanh

TensorFlow 1 version View source on GitHub

Hyperbolic Tangent (tanh) activation function.

For example:

# Constant 1-D tensor populated with value list.
a = tf.constant([-3.0,-1.0, 0.0,1.0,3.0], dtype = tf.float32)
b = tf.keras.activations.tanh(a) #[-0.9950547,-0.7615942,
0.,0.7615942,0.9950547]

Arguments: x: Input tensor.

A tensor of same shape and dtype of input x. The tanh activation: tanh(x) = sinh(x)/cosh(x) = ((exp(x) - exp(-x))/(exp(x) + exp(-x))).