François Weber

Trying to re-up this thread. I agree with @shanalikhan : a way of serializing an AutoSKL model into ONNX would be great. Would it be possible to reopen this issue...

Hi @eddiebergman and thanks for this quick answer. ASKL to SKL conversion is a really good news. IMHO as soon as a pure-SKL object is obtained, `sklearn-onnx` will be enough...

Yep, I'll definitely try to capture logs. I'll keep you posted

@lastzero my configuration is already set to "originals only". Yes @graciousgrey I confirm that selecting every pics within an album's share leads to a "normal" download. I would suggest having...

Hi @KosmosisDire , Thanks for this super neat feature :) This is perfect to send a one-file version of a note. However I had something even more compact in mind....

Okay my mistake, I was still on `1.9-1b`. Yes now it's exactly what I thought about 🎉 Thanks for reaching back. However, I notice a huge difference in terms of...

Thank you for the details. Apart of the size it's perfect for my usecase (sharing a bunch of notes with other people not familiar with MD) :) Thanks !

Thank you @NProkoptsev for adding this to LGBM :) Can't wait to to have it also in the scklearn-like API. The documentation has very few information about this new feature....

Hi, bellow is a schema that provokes the problem (@parisni please validate). Let me mention that Vespa is happy with this config file and I'm able to feed/query a Vespa...

OK thanks a lot, @jameslamb - the code is running now. However I must have done mistakes in it because the training's behavior is wrong. I give up for now...