shapviz 0.9.3

sv_dependence(): Control over automatic color feature selection

How is the color feature selected, anyway?

If no SHAP interaction values are available, by default, the color feature v' is selected by the heuristic potential_interaction(), which works as follows:

  1. If the feature v (the on the x-axis) is numeric, it is binned into nbins bins.
  2. Per bin, the SHAP values of v are regressed onto v' and the R-squared is calculated. Rows with missing v' are discarded.
  3. The R-squared are averaged over bins, weighted by the number of non-missing v' values.

This measures how much variability in the SHAP values of v is explained by v', after accounting for v.

We have introduced four parameters to control the heuristic. Their defaults are in line with the old behaviour.

If SHAP interaction values are available, these parameters have no effect. In sv_dependence() they are called ih_nbin etc.

This partly implements the ideas in #119 of Roel Verbelen, thanks a lot for your patient explanations!

Further plans?

We will continue to experiment with the defaults, which might change in the future. A good alternative to the current (naive) defaults could be:

Other user-visible changes

Small changes

Bug fixes

shapviz 0.9.2

User-visible changes

New features

Other changes

shapviz 0.9.1

New features

Maintenance

shapviz 0.9.0

New features

Documentation

API improvements

shapviz 0.8.0

API improvement

Documentation

shapviz 0.7.0

Milestone: Working with multiple ‘shapviz’ objects

Sometimes, you will find it necessary to work with several “shapviz” objects at the same time:

To simplify the workflow, {shapviz} introduces the “mshapviz” object (“m” like “multi”). You can create it in different ways:

The sv_*() functions use the {patchwork} package to glue the individual plots together.

See the new vignette for more info and specific examples.

Other new features

Maintenance

Other changes

Bug fixes

shapviz 0.6.0

Change in defaults

shapviz 0.5.0

Major improvement: SHAP interaction values

User visible changes

Compatibility with “ggplot2”

Technical changes

shapviz 0.4.1

New functionality

shapviz 0.4.0

Removed dependencies

The following dependencies have been removed:

Changes in sv_importance()

Improvement with Plotly

shapviz 0.3.0

Less picky interface

Removed (according to depreciation cycle)

Minor changes

shapviz 0.2.2

Bug fix

Minor improvements

shapviz 0.2.1

Major improvements

Minor changes

Maintainance

shapviz 0.2.0

Major improvements

Major rework of sv_importance()

The calculations behind sv_importance() are unchanged, but defaults and some plot aspects have been reworked.

Minor changes

shapviz 0.1.0

This is the initial CRAN release.