FXFloatFunctional — PyTorch 2.7 documentation (original) (raw)

class torch.ao.nn.quantized.FXFloatFunctional(*args, **kwargs)[source][source]¶

module to replace FloatFunctional module before FX graph mode quantization, since activation_post_process will be inserted in top level module directly

Valid operation names:

• add
• cat
• mul
• add_relu
• add_scalar
• mul_scalar