Aim 2.4.0 — XGBoost Integration and Confidence Interval Aggregation

Gev Sogomonian
May 18 · 2 min read

Aim 2.4.0 is out! Thanks to the community for feedback and support on our journey towards democratizing AI dev tools.

Check out the updated Aim at play.aimstack.io.

For this release, there have been lots of performance updates and small tweaks to the UI. Below are the two highlight features of Aim 2.4.0. Please see the full list of changes here.

XGBoost Integration

Now aim.callback is available that exports AimCallback to be passed to the xgb.train as a callback to log the experiments.

Check out this blogpost for more details on how to integrate Aim to your XGBoost code.

Confidence Interval as the aggregation method

Now you can use confidence intervals for aggregation. This feature has been highly-requested and completes the list of necessary aggregation methods. If you have outlier metrics in your group, confidence interval will show logical aggregation and get better comparison of the groups.

Learn More

Aim is on a mission to democratize AI dev tools.

We have been incredibly lucky to get help and contributions from the amazing Aim community. It’s humbling and inspiring.

Try out Aim, join the Aim community, share your feedback, open issues for new features, bugs.

And don’t forget to leave Aim a star on GitHub for support 🙌.

AimStack

A super-easy way to record, search and compare 1000s of AI training runs.