You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
A user recently ran into an issue when trying to upload data that it was erroring out with a "The vectors in bvecs should be unit (The tolerance can be modified as an input parameter) validation failed" error.
The validator loads the bvecs and bvals files using dipy's gradient_table_from_bvals_bvecs function, which invokes a check to see if the bvectors are normalized (see normalization check.
Specifically, the bvectors need to be constrained between -1 and 1, while some users data might have bvectors with values of 100 or greater.
DMRIPrep has a simple function for normalizing the bvectors (see normalize_vectors. We could potentially implement this upon validation as another check, or we can just better alert users on to how to fix this for their own data prior to upload.
I've written a simple python script to do this and output the normalized bvecs with a '_norm' suffix. See this repository for the script: normalize_bvectors.
This issue is not intended to necessarily be fixed or corrected within the validator, but can serve as a place for people to search for an answer if they run into this issue with their own data.
The text was updated successfully, but these errors were encountered:
Hello,
A user recently ran into an issue when trying to upload data that it was erroring out with a "The vectors in bvecs should be unit (The tolerance can be modified as an input parameter) validation failed" error.
The validator loads the bvecs and bvals files using dipy's gradient_table_from_bvals_bvecs function, which invokes a check to see if the bvectors are normalized (see normalization check.
Specifically, the bvectors need to be constrained between -1 and 1, while some users data might have bvectors with values of 100 or greater.
DMRIPrep has a simple function for normalizing the bvectors (see normalize_vectors. We could potentially implement this upon validation as another check, or we can just better alert users on to how to fix this for their own data prior to upload.
I've written a simple python script to do this and output the normalized bvecs with a '_norm' suffix. See this repository for the script: normalize_bvectors.
This issue is not intended to necessarily be fixed or corrected within the validator, but can serve as a place for people to search for an answer if they run into this issue with their own data.
The text was updated successfully, but these errors were encountered: