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.

Sign up for GitHub

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

Jump to bottom

Add SHAP explanations to models #75

Closed
pplonski opened this issue Apr 24, 2020 · 6 comments
Closed

Add SHAP explanations to models #75

pplonski opened this issue Apr 24, 2020 · 6 comments
Assignees
@pplonski
Labels
enhancement New feature or request
Projects
mljar-supervised
Milestone
version 0.3.0

Comments

@pplonski
Copy link
Contributor

pplonski commented Apr 24, 2020 •
edited
Loading

Please add

  • SHAP summary plot
  • SHAP dependence plots
  • SHAP decision plots for top-10 best predictions and top-10 worst prediction (the later might be more important)
@pplonski pplonski added the enhancement New feature or request label Apr 24, 2020
@pplonski pplonski self-assigned this Apr 29, 2020
@pplonski pplonski added this to the version 0.3.0 milestone Apr 29, 2020
@pplonski pplonski added this to To do in mljar-supervised Apr 29, 2020
@pplonski pplonski pinned this issue Apr 29, 2020
@pplonski pplonski moved this from To do to In progress in mljar-supervised Apr 29, 2020
@pplonski
Copy link
Contributor Author

Please be careful with shap values shape. Take a look at shap/shap#526

pplonski added a commit that referenced this issue May 1, 2020
@pplonski
add SHAP importance plots (#75)
d18505b
pplonski added a commit that referenced this issue May 4, 2020
@pplonski
SHAP explanations (#75)
bec45a3
@pplonski pplonski moved this from In progress to Done in mljar-supervised May 4, 2020
@pplonski
Copy link
Contributor Author

pplonski commented May 4, 2020

Done. Tests will be added with this issue #69

@pplonski pplonski closed this as completed May 4, 2020
@pplonski pplonski unpinned this issue May 4, 2020
@Tonywhitemin
Copy link

Hi @pplonski,
It's me again (laugh), sorry for that but I am little confused now...
Image below is the result of "SHAP feature important".
I don't know how to explain why the trend of Learner1 is almost reversed with SHAP value?
Could you help? Thanks!
image

@pplonski
Copy link
Contributor Author

Hi, I have no idea, maybe try to ask on SHAP github repo?

@Tonywhitemin
Copy link

Ok, thanks!👍

@StrangeNPC
Copy link

Hi @Tonywhitemin , did you ever find out why?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@pplonski pplonski

Labels
enhancement New feature or request
Projects
mljar-supervised
  
Done
Milestone
version 0.3.0
Development

No branches or pull requests
3 participants
@pplonski @Tonywhitemin @StrangeNPC