Sparsemax activation function.

For each batch i, and class j, sparsemax activation function is defined as:

sparsemax(x)[i, j] = max(x[i, j] - (x[i, :]), 0).

Usage

activation_sparsemax(x, axis = -1L)

Arguments

x

Input tensor.

axis

int, axis along which the sparsemax operation is applied. (1-based)

Value

A tensor, output of sparsemax transformation. Has the same type and shape as x.

Reference

• Martins et.al., 2016

See also

Other activations:
activation_celu()
activation_elu()
activation_exponential()
activation_gelu()
activation_glu()
activation_hard_shrink()
activation_hard_sigmoid()
activation_hard_tanh()
activation_leaky_relu()
activation_linear()
activation_log_sigmoid()
activation_log_softmax()
activation_mish()
activation_relu()
activation_relu6()
activation_selu()
activation_sigmoid()
activation_silu()
activation_soft_shrink()
activation_softmax()
activation_softplus()
activation_softsign()
activation_sparse_plus()
activation_squareplus()
activation_tanh()
activation_tanh_shrink()
activation_threshold()