LazyLinear(out_features, bias=True, device=None, dtype=None)¶
torch.nn.Linearmodule where in_features is inferred.
In this module, the weight and bias are of
torch.nn.UninitializedParameterclass. They will be initialized after the first call to
forwardis done and the module will become a regular
in_featuresargument of the
Linearis inferred from the
torch.nn.modules.lazy.LazyModuleMixinfor further documentation on lazy modules and their limitations.
out_features – size of each output sample
bias – If set to
False, the layer will not learn an additive bias. Default:
~LazyLinear.weight – the learnable weights of the module of shape . The values are initialized from , where
~LazyLinear.bias – the learnable bias of the module of shape . If
True, the values are initialized from where