I have tried to make a 3D convolutional layer with the following parameters:
padding='same' and groups='in_channels' and l get an AssertionError. Either with
tltorch.FactorizedConv() or tltorch.FactorizedConv.from_conv()
It seems that padding='same' with groups is not supported in tensorly.
Thank you for your help.
I have tried to make a 3D convolutional layer with the following parameters:
padding='same'andgroups='in_channels'and l get anAssertionError.Either withtltorch.FactorizedConv()ortltorch.FactorizedConv.from_conv()It seems that
padding='same'withgroupsis not supported in tensorly.padding='same'in pytorch pytorch doc on paddinggroups= in_channelsin pytorch groups in convThank you for your help.