Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Visualize single prediction #29

Closed
mayer79 opened this issue May 16, 2022 · 3 comments
Closed

Visualize single prediction #29

mayer79 opened this issue May 16, 2022 · 3 comments

Comments

@mayer79
Copy link
Contributor

mayer79 commented May 16, 2022

I am unsure: do we already have a plot that shows SHAP decomposition for a single observation? Something like

If not: we might try to import such graphs from other R packages or write our own version(?)

@liuyanguu
Copy link
Owner

We kind of have the force plot using (shap.plot.force_plot) but there is no decomposition for a single observation. That sounds a good idea to me. :)

@mayer79
Copy link
Contributor Author

mayer79 commented May 17, 2022

The package "waterfalls" has minimal dependencies (ggplot2 and grDevices), so this is a good candidate for waterfall plots: (https://cran.r-project.org/web/packages/waterfalls/index.html)

Is it possible to modify shap.plot.force_plot to work for single observations? Or is it quite different from the one in Python (see figure above)?

@liuyanguu
Copy link
Owner

I kind of feel that we shall use a new function for it. The original shap.plot.force_plot was too complicated. Sounds very interesting I will further check the waterfalls package ...

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants