Weisu Yin

Autopilot Advertisement

1

*Issue #, if available:* *Description of changes:* **Screenshots below are rendered locally for simplicity. Please check the built website after CI is finished** * Added banner to advertise Autopilot and...

Example of how torch is doing it:

TabularPredictor will drop columns sometime, and this would bring trouble to model deployment when user omits the header info to minimize payload size. In such case, we would need to...

Segmentation Fault when Importing AutoGluon

3

We have noticed that a specific order of import AutoGluon modules will result in seg fault. For example, ```python import autogluon.text # same applies to autogluon.multimodal from autogluon.vision import ImagePredictor...

Use `higher_is_better` from `torch_metrics` instead of manual `MIN` and `MAX`

1

@gradientsky pointed out it might be better to use `higher_is_better` from [torchmetrics](https://torchmetrics.readthedocs.io/en/stable/pages/overview.html#metrics-and-hyperparameter-optimization) @sxjscience @zhiqiangdon @bryanyzhu we can discuss here to share opinions.

[Security] Support for torch lightning 1.6 and future support

5

Currently the master branch supports TL 1.5. What's the plan and timeline regarding TL 1.6? Also, we want to utilize distributed HPO with each trial being distributed itself, and found...

[Tune] PBT/PB2 doesn't work correctly with Ray Lightning

3

When using PBT/PB2, I received the following error: ```shell RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu! ``` This...

[Tune] Ray Tune gives incorrect warning about not using GPUs when used with Ray Lightning

4

When trained on gpu, I see the following warning: ```shell 2022-05-03 00:02:20,033 WARNING tune.py:637 -- Tune detects GPUs, but no trials are using GPUs. To enable trials to use GPUs,...

Not detecting GPU within trials when HPO

5

I tried to use ray lightning + ray tune to do distributed HPO and found GPUs are not available within the trial even when I set `use_gpu` to be True...

- [ ] I have checked that this bug exists on the latest stable version of AutoGluon - [ ] and/or I have checked that this bug exists on the...