hard_sigmoid¶

paddle.fluid.layers.nn. hard_sigmoid ( x, slope=0.2, offset=0.5, name=None ) [source]

HardSigmoid Activation Operator.

A 3-part piecewise linear approximation of sigmoid(https://arxiv.org/abs/1603.00391), which is much faster than sigmoid.

\(out = \max(0, \min(1, slope * x + offset))\)

Parameters

• x (Variable) – An N-D Tensor with data type float32, float64.

• slope (float, optional) – The slope of the linear approximation of sigmoid. Its value MUST BE positive. Default is 0.2.

• offset (float, optional) – The offset of the linear approximation of sigmoid. Default is 0.5.

• name (str, optional) – The default value is None. Normally there is no need for user to set this property. For more information, please refer to Name

Returns

A Tensor with the same shape as input.

Return type

Variable

Examples

import paddle.fluid as fluid
import paddle
paddle.enable_static()

data = fluid.layers.fill_constant(shape=[3, 2], value=0.5, dtype='float32') # [[0.5, 0.5], [0.5, 0.5], [0.5, 0.5]]
result = fluid.layers.hard_sigmoid(data) # [[0.6, 0.6], [0.6, 0.6], [0.6, 0.6]]