B02-TR02 · official_board_snapshot
promoter_all
promoter_all
Protocol fingerprint
PF-v1-CF0E82D1E88AEE10BDEED1CA
The result chunk loads only after both the metric and protocol-metric fingerprint are selected.
- Split
- official tenfold cross-validation test folds
- Aggregation
- arithmetic mean across 10 exact values in the official MCC array
- External data
- model-specific pretraining corpora allowed; no extra downstream labels stated
- Ensemble / best-of-N
- single model configuration summarized across cross-validation folds · 1
- Critical unknowns
- external_data_policy;inference_budget
Interactive results
Filter and compare inside one fingerprint
Choose a metric and fingerprint to load results.
Top 50 rank-eligible results
Current fingerprint and filters only.
Same-fingerprint comparison
| Participant | Configuration | Score | Rank | Claim |
|---|
| Compare | Participant / configuration | Score | Ranks | Baseline / claim | Evidence & settings |
|---|
Complete protocol fields
- canonical dataset ids
- nt_promoter_all
- data version
- official HF leaderboard CSV sha256:a93c1e722f6b15312818903dc048207845598925889f415caaf61e3af7d60ca6
- label visibility
- training-fold labels visible; held-out fold labels unavailable to each fold model
- input modality
- nucleotide sequence
- preprocessing
- official Nucleotide Transformer downstream benchmark preprocessing
- model selection rule
- official leaderboard procedure
- inference budget
- NR
- protocol completeness
- exact