Hardtanh¶

class torch.nn.Hardtanh(min_val: float = -1.0, max_val: float = 1.0, inplace: bool = False, min_value: Optional[float] = None, max_value: Optional[float] = None)[source]¶

Applies the HardTanh function element-wise

HardTanh is defined as:

$$\text{HardTanh}(x) = \begin{cases} 1 & \text{ if } x > 1 \\ -1 & \text{ if } x < -1 \\ x & \text{ otherwise } \\ \end{cases}$$

The range of the linear region $[-1, 1]$ can be adjusted using min_val and max_val.

Parameters

• min_val – minimum value of the linear region range. Default: -1

• max_val – maximum value of the linear region range. Default: 1

• inplace – can optionally do the operation in-place. Default: False

Keyword arguments min_value and max_value have been deprecated in favor of min_val and max_val.

Shape:

• Input: $(N, *)$ where * means, any number of additional dimensions

• Output: $(N, *)$ , same shape as the input

Examples:

>>> m = nn.Hardtanh(-2, 2)
>>> input = torch.randn(2)
>>> output = m(input)