Rate this Page
★★★★★
Linear#
- classtorch.ao.nn.qat.dynamic.Linear(in_features,out_features,bias=True,qconfig=None,device=None,dtype=None)[source]#
A linear module attached with FakeQuantize modules for weight,used for dynamic quantization aware training.
We adopt the same interface astorch.nn.Linear, please seehttps://pytorch.org/docs/stable/nn.html#torch.nn.Linearfor documentation.
Similar totorch.nn.Linear, with FakeQuantize modules initialized todefault.
On this page
