Hardswish# class torch.ao.nn.quantized.Hardswish(scale, zero_point, device=None, dtype=None)[source]# This is the quantized version of Hardswish. Parameters scale – quantization scale of the output tensor zero_point – quantization zero point of the output tensor