Integrate metrics into aind-motion-correction capsule

[arielleleon]

No description provided.

[arielleleon]

Sean will create a new branch on aind-ophys-motion-correction to integrate metrics. He will test in the CodeOcean Test environment by creating a pipeline with only this capsule

blocked by #37

[arielleleon]

#37 Is now ready

[arielleleon]

@seanmcculloch will work on this today

[arielleleon]

@seanmcculloch To have this done by the end of the week.

[seanmcculloch]

Using branch 40-create-qc-json

Testing from capsule: https://codeocean.allenneuraldynamics.org/capsule/9347369/tree

Current Status:

• Modeling the QC.json for motion_correction after module: FOV Quality and Motion Correction (example)
• quality_control.py creates the quality_control.json
• Plots used:
  • <plane_id>_registration_summary.png
  • combines max and average projections into <plane_id>_combined_projections.png

Next Steps:

1. Need to run on existing data, and manually upload the created quality_control.json to docDB
   a. Example capsule here for pushing to docDB: see
   "Alternate Workflow" at the end of run_capsule.py (As-is this capsule writes to the dev docDB)
   b. Or, write quality_control.json to the data asset in S3, then wait for sync to docDB (every 4hrs)
   Note: Ensure filepath plots are accurate to their final locations in the data asset. The 'reference' field of each QC metric should be a relative path from the root of the data asset.(This would be done at the later metadata aggregtion step but is needed now to be able to view in the QC portal.