Repository navigation
optimize loading and application of AI model for foward HB tracking - #1563
Merged
Merged
Conversation
tongtongcao
requested review from
baltzell,
hauenst and
zieglerv
as code owners
November 3, 2026 02:44
Collaborator
|
Note, might as well just have a field scale for the model, instead of "isInbending" and the extra logic. |
baltzell
approved these changes
Nov 6, 2026
Collaborator
Author
isInbendingHBTSEModel points out that the model was trained using training samples from inbending or outbending data. It will tell if predicted sign of Q needs to be flipped, depending on scale factor for magnetic field. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Originally, two separate models are applied for HB tracking: One for inbending, and the other for outbending.
But actually, a model trained on inbending samples also works well for outbending reconstruction by simply flipping the sign of the predicted Q parameter, and vice versa.
Besides, normalization values for features and output states are previously hard-coded.
With this update, these values are loaded from separate statistics files, which are generated during model training based on the training samples. The files are added through the networks by PR https://code.jlab.org/hallb/clas12/coatjava/networks/-/merge_requests/14
This PR prepares for the integration and application of dedicated models for RGL, and will be convenient for application of future models.