PROC GRADBOOST AUTOTUNE: Partition-level Gini for all evaluated configs without refitting?

Zaid_Sanchez — Mon, 02 Feb 2026 19:58:31 GMT

I’m tuning a binary model with PROC GRADBOOST + AUTOTUNE and PARTITION (train/validation/test). The HISTORYTABLE= captures one objective per evaluation (Gini on validation or the k-fold average), but I need Gini per partition (train/valid/test) for every evaluated configuration. Right now, my workaround is to refit each configuration, score, and run PROC ASSESS BY _PartInd_ to compute AUC → Gini, which is costly.
Questions:

Is there a built-in way to get partition-level metrics (train/valid/test) for all evaluated configurations without refitting each one?
Is there a supported pattern to persist each candidate model during tuning (not just the champion) so I can score later and compute partition metrics without retraining?
Any roadmap to add partition-level metrics (or an option to save all intermediate models) to AUTOTUNE outputs?

From the docs, AUTOTUNE records a single objective per evaluation (validation/k-fold) in the history and doesn’t expose per-partition metrics by default.

Re: PROC GRADBOOST AUTOTUNE: Partition-level Gini for all evaluated configs without refitting?

sbxkoenk — Thu, 05 Feb 2026 23:26:23 GMT

Hello,

As far as I know, it's indeed not possible to get the tuner objective function (GINI for example or Misclassification Error Percentage) for all partitions. The objective metric you see (in autotune historytable=) is the one coming from validation calculations.

Remark that a partition is always used by Autotune, specifically to avoid overfitting. This partition (separate table for scoring/validation) is either supplied by the user or created internally by Autotune.

Here's where you can make product-suggestions for the roadmap:
https://communities.sas.com/t5/SAS-Product-Suggestions/idb-p/product-suggestions
(previously known as SASware Ballot)
You can also open a Technical Support ticket instead and ask for this. (https://support.sas.com/en/technical-support.html)

Some references:

Automated Hyperparameter Tuning for Effective Machine Learning (SAS Global Forum 2017 paper)
Boost ML accuracy with hyperparameter tuning (with a fun twist) - SAS Users (By Stu Sztukowski on June 13, 2025)
The afterparty: Hyperparameter autotuning revisited - SAS Users (By Stu Sztukowski on SAS Users December 17, 2025)
SAS Help Center: AUTOTUNE Statement (GRADBOOST Procedure)
SAS Help Center: AUTOTUNE Statement (Machine Learning Procedures -- Shared Concepts)

Ciao,

Koen

topic PROC GRADBOOST AUTOTUNE: Partition-level Gini for all evaluated configs without refitting? in SAS Data Science

PROC GRADBOOST AUTOTUNE: Partition-level Gini for all evaluated configs without refitting?

Re: PROC GRADBOOST AUTOTUNE: Partition-level Gini for all evaluated configs without refitting?