Tomáš Frýda

@wendycwong The comment you wrote was for functionality implemented in another PR on which this PR was based on. The former, which dealt with time allocation, is already merged. This...

@magrenimish Thank you for creating this issue and bringing this to our attention. AutoML should have failed with a nicer message, e.g., `No model was trained.`. GBM requires more data...

@magrenimish that's basically what should happen. AutoML doesn't want to know about underlying constraints of individual models so first each model runs its parameter/training data validation logic and if that...

@magrenimish you can just ignore the warning. When I run your code, I can still get the automl to train and it looks some GBMs have parameters that enable training...

Hi @MoonCapture, 1) Calculating SHAP for stacked ensemble has a big memory requirement - it calculates baseline SHAP for every basemodel and the metalearner. SHAP for non-tree models also requires...

@MoonCapture I believe you can specify the `background_frame` in order to get the SHAP plots for all the models that support it, e.g., `h2o_01.explain(df_test, background_frame=df_train)` should produce the SHAP plot...

@MoonCapture You can get the data for PDP by setting plot=False, e.g., `model.partial_plot(df, cols=.., plot=False)` then you can use any library to draw the plot. Customization of the h2o plots...

@MoonCapture `pd_plot` uses the data from `partial_plot` and for varimp you can use `varimp()`.

> I prefer to merge this change after the major release. The release pipeline has been moved from Jenkins master to regular workers. I would like to reduce a number...

Thank you for taking time to pinpoint the issue. Unfortunately, I don't have Windows machine so I have just 2 untested hypotheses: (1) unexpected character in the path, (2) two...