LazyConvTranspose3d#
- class torch.nn.LazyConvTranspose3d(out_channels, kernel_size, stride=1, padding=0, output_padding=0, groups=1, bias=True, dilation=1, padding_mode='zeros', device=None, dtype=None)[source]#
一个具有懒惰初始化
in_channels
参数的torch.nn.ConvTranspose3d
模块。ConvTranspose3d
的in_channels
参数将从input.size(1)
推断。将惰性初始化的属性是 weight 和 bias。有关惰性模块及其限制的更多文档,请参阅
torch.nn.modules.lazy.LazyModuleMixin
。- 参数
- cls_to_become[source]#
alias of
ConvTranspose3d