Is there an implementation of MDLSTM?

I was working on an implementation of a MDLSTM as described in this paper but I can’t have it predict anything.
Therefore I wanted to compare my implementation to an existing functioning one but couldn’t find any.
Does any one of you guys where I could find one ? Does pytorch plan to provide such a component in the future ?
Thanks for your help

1 Like

I would also find this useful.

This package has " * layers.BDHW_LSTM: an MDLSTM variant that is a drop-in replacement for a Conv2d layer" but I haven’t looked at the code.

Thanks a lot for the information. I’ll look into it.