Calling forward() of ConvTranspose2d after putting the module into Sequential gave an exception

Hello,

I created ConvTranspose2d and put it in Sequential. Calling forward() afterwards gave me an exception. It seems like asserting the argument has an issue with that since ConvTranspose2d.forward() requires two arguments while I put just a tensor in the Sequential.forward().

Would you please take a look at this and let me know what I should do about this?

Can you post your code along with the exception?

@ldw This is a known issue in libtorch 1.4 and will be fixed in 1.5 (which is slated to be released in April) :smiley:

Related PR: https://github.com/pytorch/pytorch/pull/33027

1 Like