bgc-setnet / source /data /manifests /data_quality.json
whiteh4t's picture
Release final BGC retrieval checkpoints and model card
c87881a verified
Raw
History Blame Contribute Delete
1.17 kB
{
"atlas_file_sha256": "4a30a45f95576602200c26bc4de4db979d2d21b717aa078d12aff3b5599ef0cd",
"embedding_coverage": 0.9998074383802817,
"embeddings_file_sha256": "2fd3c7b220c0cb26734505fef637811fa48bfb97ce55a1d7ede56e8451310d47",
"missing_embedding_rows": 35,
"model_input_whitelist": [
"gene_embeddings",
"padding_mask",
"relative_positions"
],
"position_source": "ordered community_atlas.protein_ids",
"schema_version": 1,
"split": {
"by_split": {
"test": {
"bgcs": 438,
"groups": 21,
"label_tiers": {
"silver": 438
}
},
"train": {
"bgcs": 4211,
"groups": 98,
"label_tiers": {
"silver": 4211
}
},
"validation": {
"bgcs": 676,
"groups": 21,
"label_tiers": {
"silver": 676
}
}
},
"schema_version": 1,
"split_sha256": "197c97bbc3bec139166f5ad367508853139295f9b984ccf6b8ad360038bd6d0b"
},
"split_file_sha256": "dc26fae17e54fd2ad41a9e10353b3da3e0aacf3144b64f6ee62e8341b4360555",
"usable_bgcs": 5325,
"usable_gene_rows": 181725,
"zero_gene_fallback": false
}