Michael Mayer comments

Results 161 comments of


                                            Michael Mayer

[R] Fix xgb.cv() for AFT models

We *could* add more arguments to `xgb.cv()`. I was actually thinking in the other direction: let the function work only with xgb.DMatrix input and remove the "label" argument. Why? The...

[R] Fix xgb.cv() for AFT models

Closed in favour of #10031

Roadmap for new R interface

@david-cortes: fantastic road map, thank you so much. Unfortunately, you have spotted my biggest weakness! For the C part, we might ask the data.table team. For the C++ part, Dirk...

Roadmap for new R interface

Good idea. I even remember this issue from somewhere.

change in R 4.2 disallows a comparison used in xgboost()

Can you provide a small reproducible example please?

[mrg] ENH Add custom_range argument for partial dependence - version 2

@thomasjpfan : this functionality is very high up on my wish list, e.g., to reorder the categories of a discrete feature into a natural order. How can we bring this...

[mrg] ENH Add custom_range argument for partial dependence - version 2

Removed (wrong) comment of mine regarding Friedman's H statistic. To calculate it, we would need to be able to pass a custom grid also for the 2-dimensional case.

partial_dependence() computes conditional partial dependence

@lorentzenchr : the two methods are different, and they estimate different things. (Just like TreeSHAP and permutation SHAP. ) I think it would be good to mention this in the...

ENH Add Friedman's H-squared

Note: The current output API can be easily converted to a pandas dataframe and plotted, just like this: ```py import seaborn as sns import pandas as pd from sklearn.compose import...

ENH Add Friedman's H-squared

@amueller @glemaitre @adrinjalali @lorentzenchr Let me know if you see anything that is missing or should be improved.