-
Notifications
You must be signed in to change notification settings - Fork 34
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
monotonic constraints not working in xgboost_ray #273
Comments
XGBoost Ray Monotonic Constraints.ipynb.zip attached the notebook in .ipynb format for convenience |
Hey @heyitsmui, I cannot reproduce this. I get the exact same plot from both xgboost and xgboost-ray (with 1 or 2 workers). I used your notebook and edited the saving of the second plot so it doesn't override the previous plot. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Was chatting with @krfricke about this in recent weeks, the monotonic constraints parameter doesn't seem to be working on
xgboost_ray
.Steps to reproduce: Provided a notebook here which compares an open source xgboost model vs. open source xgboost_ray trained model, with monotonicity enabled for certain features.
debug_xgboost_ray_monotonic_constraints (1).pdf
What do we expect: Monotonicity should be enforced. EG in the notebook above,
In this plot, the first subplot should be strictly monotonically increasing as the x-axis increases but we see a dip at the beginning when x ~= 0.5ish
The text was updated successfully, but these errors were encountered: