-
Notifications
You must be signed in to change notification settings - Fork 39
xgb: infer metric data names from evals and deprecate metric_data
#587
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
Conversation
Codecov ReportPatch coverage:
Additional details and impacted files@@ Coverage Diff @@
## main #587 +/- ##
==========================================
+ Coverage 89.65% 89.71% +0.05%
==========================================
Files 43 43
Lines 2938 2955 +17
Branches 242 245 +3
==========================================
+ Hits 2634 2651 +17
Misses 264 264
Partials 40 40
☔ View full report in Codecov by Sentry. |
|
Or should we remove the Alternatively, |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hi @sisp thanks for the contribution!
I took a look and I think what makes sense is to just go ahead and drop metric_data , as you suggested in #587 (comment)
|
@daavoo Shouldn't we provide at least a grace period until the breaking change takes effect, still violating SemVer but only after, e.g., 3 minor releases? We could allow |
Yes, sorry for the lack of clarity, I meant the option in your last paragraph |
|
Great! I'll update the PR tomorrow. |
evals and deprecate metric_data
|
@daavoo I've updated the PR as we discussed and resolved merge conflicts with |
Co-authored-by: David de la Iglesia Castro <daviddelaiglesiacastro@gmail.com>
for more information, see https://pre-commit.ci
daavoo
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thank you for contributing and addressing all comments!
|
You're welcome! Thanks for the thorough review and your responsiveness! 🙏 🙇 |
I've added support for logging metrics for multiple data sets/splits (e.g. train and test/eval) with XGBoost. This is consistent with other supported frameworks, e.g. Keras. It's also a common use case to compare the metrics on multiple data sets/splits, e.g. to check overfitting.
This addition is backwards compatible.
I'll create a PR to the docs project once this addition has been approved.
❗ I have followed the Contributing to DVCLive
guide.
📖 If this PR requires documentation updates, I have created a separate PR (or issue, at least) in dvc.org and linked it here.