HirakoSan/bio-lens

🌿 iNaturalist Bronze Dataset (Research-Grade, Deduplicated)

Description

This dataset contains research-grade observations from iNaturalist, processed through a bronze-layer pipeline that includes: It's a curated dataset for biodiversity analysis based on community sourced observations, contains millions of images thus can be used for biology model training.

The intent of this data is to train specialist vision models capable of identifying species with high accuracy.

• Research-grade only observations (community-verified)
• One photo per observation (deduplicated by observation_uuid, keeping the first photo)
• AVIF encoded images stored as binary in Parquet
• ~5TB total size across over 41,316 shards

Note: This dataset started as all photos, but was tuned to filter downloaded images to only research-grade observations with 1 photo per observation. This prioritizes observation diversity over multiple (often near-identical) photos of the same specimen.

Data Composition

Metric	Count
Total bronze rows (photos)	> 400M
Research-grade photos (with images)	> 140M
Research-grade observations	> 140M
Unique taxa	> 500K
Shards	41,316 Parquet files
Photo/Obs count per shard	10,000
Shard size	~500MB each

Top Families

Family	Photos
Asteraceae (daisies/sunflowers)	11.7M
Fabaceae (legumes)	6.4M
Nymphalidae (brush-footed butterflies)	5.6M
Anatidae (ducks/geese)	4.8M
Rosaceae (roses)	3.6M
Apidae (bees)	3.4M
Lamiaceae (mint family)	3.0M
Poaceae (grasses)	2.9M
Accipitridae (hawks/eagles)	2.9M
Orchidaceae (orchids)	2.7M

File Structure

data/train/
├── 0000.parquet
├── 0001.parquet
├── ...
└── 41316.parquet.parquet

Schema

Column	Type	Description
photo_id	int64	Unique photo identifier
observation_uuid	string	UUID (deduplicated, one per observation)
obs_taxon_id	int64	iNaturalist taxon ID
taxon_name	string	Scientific name
ancestry	string	Taxonomic lineage path
rank	string	species, genus, family, etc.
image	binary	AVIF encoded image (NULL for non-research-grade)
license	string	Photo license
obs_quality_grade	string	"research", "casual", or "needs_id"
obs_latitude	double	Observation latitude
obs_longitude	double	Observation longitude

Note: Only rows with obs_quality_grade == "research" have the image column populated. Other quality grades have NULL images.

Image Format

• Format: AVIF (AV1 Image File Format)
• Quality: 75
• Dimensions: Original (preserved)
• Average size: ~15-30 KB
• Metadata: Stripped (EXIF/XMP removed)

Licenses

Photos include licenses as selected by observers: CC0, CC-BY, CC-BY-NC, CC-BY-SA, CC-BY-NC-SA, CC-BY-ND, CC-BY-NC-ND.

⚠️ Non-commercial research use only

Usage

import polars as pl

# Load only research-grade images
df = pl.read_parquet("data/train/*.parquet")
research_df = df.filter(pl.col("obs_quality_grade") == "research")

# Filter by family
beetles = research_df.filter(pl.col("family") == "Carabidae")

# Access image
row = research_df.first()
image_bytes = row['image']

Processing Pipeline

This dataset was created by:

1. Joining photos, observations, and taxonomy
2. Writing all photos (research + casual + needs_id) to bronze parquet
3. Downloading and converting images only for research-grade observations
4. Storing images in the image column (NULL for non-research-grade)
5. Deduplicating to one photo per observation
6. Sharding into ~500MB Parquet files

Citation

@misc{bio-lens-bronze,
  author = {iNaturalist Contributors},
  title = {iNaturalist Bronze Dataset (Research-Grade, Deduplicated)},
  year = {2026},
  publisher = {Hugging Face},
  howpublished = {\url{https://huggingface.co/datasets/HirakoSan/bio-lens}}
}

Notes

• The dataset can be further augmented, it would represent over ~12 TB total
• Only ~142.8M rows have images (research-grade)
• The remaining rows (casual/needs_id) have NULL image columns and can be filtered out
• One photo per observation favours observation diversity