Provenance of CanonBeetles.csv

[hlapp]

@SMRayeed can you clarify the provenance of the CanonBeetles.csv file that is mentioned by absolute pathname (on the OSC cluster) in one of the notebooks?

[SMRayeed]

The CanonBeetles.csv (and similarly SwiftBeetles.csv) file is derived from our Hawaii-2025 beetle imaging photo log maintained in Google Sheets: link here; located in the shared drive BeetleIntake link here

The spreadsheet is used to track specimen imaging metadata. These CSV files were generated by exporting the relevant entries from the photo log and then running Grounding-DINO to localize bounding box coordinates for each specimen image.

[hlapp]

Thanks @SMRayeed. The Google sheets are not openly accessible; I've requested access if that needs to remain so. It also looks like these sheets then need to be part of either the Hawaii-beetles dataset (is this what you mean by Hawaii-2025 beetle imaging?) or this repo?

[egrace479]

These CSVs (or the means to re-create them) need to be included for re-use.

Additionally, issue #5, regarding the alignment of filenames is also needed for others to be able to pull the data and re-use this pipeline. As an explicit example, the Benchmarking/Beetle-PUUM/BeIT-species.csv has filenames/paths that do not match those used in the published dataset. These must be mapped between the IMG_####.jpg/uuid.png and the IMG_<id>_specimen_<number>_<taxonID>_<individualID>.png filenames used for Hawaii Beetles. See #5, for more on this issue as it is the same for the other data sources.