Nick Erickson
Nick Erickson
@serhiyandrejev we are currently migrating all vision logic to autogluon.multimodal . Can you try fitting your dataset with MultimodalPredictor? It might be that vision module doesn't pickup GPU properly on...
@yxxan It does not by default, however in v0.3 we added support for class imbalance handling via the `sample_weight` parameter that can be specified during predictor init: https://auto.gluon.ai/stable/api/autogluon.task.html#module-0 ``` predictor...
@rxjx Only for Tabular at present.
@Jalagarto We don't currently support customized losses, however if you were to implement support for it on a model by model basis, we are happy to accept contributions.
Added in https://github.com/awslabs/autogluon/pull/2071
Unfortunately I was never able to reproduce this issue. It may have been fixed in v0.5 or the upcoming v0.6 release. If you can provide a reproducible example using latest...
@gradientsky @yinweisu @gidler This PR is ready to review. Note that the specifics of the cascade implementation are primarily from a logical correctness POV. There is a future PR that...
> LGTM!! One tiny comment is that consider adding more comments explaining the logic especially when nested index is happening in the code. Would help a lot to understand the...
@navinp0304 This bug has been fixed in mainline, and will be available in the upcoming v0.6 release. Thanks again for reporting!
I just ran into this issue as well, and it was quite counterintuitive the root cause. It would be very nice if this worked. If I have a branch of...