Chi Wang
Chi Wang
> Can i work on this one? Sure. Assigned to you.
> The issue was already resolved Was it? I haven't seen a PR fixing it.
> Sorry i found it, in which repository can i find the code to commit the change? The current repo, under website/docs/. You could also click "edit this page" from...
Have you tried to use pandas dataframe? The categorical features are supposed to be recognizable by flaml when using dataframe. And you don't need to pass that info to flaml....
`gpu_per_trial` only works for xgboost and xgb_depth for 'classification' task. You can follow this example to specify `fit_kwargs_by_estimator`: https://microsoft.github.io/FLAML/docs/Use-Cases/Task-Oriented-AutoML#extra-fit-arguments
Could you share the code that works for xgboost? And the xgboost version in both cases?
Please use https://microsoft.github.io/FLAML/docs/Contribute#pre-commit to pass code format check.
The issue can be solved by commenting this line: https://github.com/microsoft/FLAML/blob/784c4c4f86e8b0c6abeb411f06b3a1b78a74980b/flaml/automl/model.py#L1513 Would you like to make a PR?
> Is it possible to summarize the observed reasons for the failures? One can run these examples and observe the chat. Then, one can try adding hints to make it...
Specifying X_val and y_val won't work when using MultiOutputRegressor. Workaround: combine training and validation data as X_train and y_train, and then use a custom splitter to split the data as...