This manifest enumerates every photo used to train, validate, and test the model. Each row links back to the original observation and photo on iNaturalist, preserving full attribution and license metadata.

Source

Model run: 20260528_104624_licensed_dinov2_l_14_reg4_518
Generated: 2026-05-29T02:17:00.691986+00:00

Headline numbers


Total photos	123,099
Unique observations	60,859
Species (classes)	161
Families	27

Splits

Split	Photos
test	18,393
train	86,085
val	18,621

Regions

Region	Photos
nh	123,099

License distribution

Every photo retains its original iNaturalist license. The model was trained under a permissive-license policy at the project level.

License	Photos
cc-by-nc	102,952
cc-by	13,971
cc0	3,570
cc-by-nc-sa	1,453
cc-by-sa	1,153

Schema

Column	Description
`label_idx`	Integer class index used by the model
`species`	Genus_epithet (e.g. `Phidippus_audax`)
`split`	`train`, `val`, or `test`
`observation_id`	iNaturalist observation ID
`photo_id`	iNaturalist photo ID
`final_license_code`	License of the photo (e.g. `cc0`, `cc-by`, `cc-by-nc`)
`license_source`	Whether the license came from the observation CSV or recovery pass
`image_attribution`	Photographer attribution string supplied by iNaturalist
`region`	`nh` or `bordering` (geographic filter at scrape time)
`family`	Taxonomic family
`inat_photo_url`	Canonical iNaturalist photo URL
`inat_observation_url`	Canonical iNaturalist observation URL

Acknowledgments

All photos were sourced from iNaturalist under their respective Creative Commons or other open licenses. Each row's image_attribution field credits the individual observer who contributed the photo. Please follow the terms of each photo's license when reusing.

Reproducibility

This manifest was produced by the project's deterministic pipeline. To regenerate from the model's bundled run directory, run pipeline/6_publish_manifest.py against the source dataset_manifest.csv.

Downloads last month: 7

Total file size:

28.6 MB