metadata
license: mit
Model Information π§¬
License: MIT
π¬ Base Model:
𧩠Task Type:
Protein-level regression
π Dataset:
- protein: Contains mutation data including the RhlA enzyme sequence and corresponding performance metrics.
- Label: The experimentally tested fitness score, representing the scaled mutation effect for each mutant.
- Source: Label derived from CAPE
π‘ Model Input Type:
Amino acid sequence; label in RhlA
π Performance (the best on test set):
Spearman's Ο: 0.862
LoRA Configuration βοΈ
- r: 8
- LoRA dropout: 0.1
- LoRA alpha: 8
- Modules to save:
["regression"]
Training Configuration ποΈ
- Optimizer:
- Class: AdamW
- Betas: (0.9, 0.98)
- Weight decay: 0.01
- Learning rate: 5e-5
- Epochs: 5
- Batch size: Adaptive