Bas Veeling

> There is also #1018 (which aims to support only one model listed in your description). I think it would be great if you @basveeling and @sschirr could combine your...

@rytilahti Thank you for the review! With a bit of delay on my side, I believe I have made most of the changes requested, outside of moving the bytestring building...

@rytilahti Thank you for the feedback! A couple of changes in this new version. - moved the recipe building to Construct (I needed to add two custom Subconstructs for this,...

Hi @rytilahti, thanks for the suggestion. I've refactored the code to use the new dir structure. It's a large PR so I'd understand if it takes some time :-). Looking...

Hi Teemu / @rytilahti, I hope all is well. I see there's still a change requested status on this PR, but I think I've addressed all the comments. Just wanted...

Hi! I'm a bit swamped right now due to conference deadlines and do not have time to look into this. I'll try and get back to you on this soon....

No, but with the legacy changes it would probably work just fine. Do note that I've pushed causal 1d convolutions to keras, but with a different api (using the padding...

Keras provides a legacy layer that should make it easy to run keras v1 code on keras v2. Feel free to post any tracebacks and errors here.

Hi all, Thanks for bringing up this issue! I'm a bit swamped right now due to conference deadlines and do not have time to look into this. If you find...

I haven't, but this would be interesting! I don't have any access to NN training hardware right now, but I'd love to see if this works.