metadata
base_model: westlake-repl/SaProt_650M_AF2
library_name: peft
Model Card for Model ID
This model is used to predict solubility of a amino acid sequence.
Task type
protein level classification
Dataset description
The dataset is from DeepSol: a deep learning framework for sequence-based protein solubility prediction. Binary label, 1 means soluble, 0 means insoluble.
Model input type
Amino acid sequence
Performance
test_acc: 0.74
LoRA config
lora_dropout: 0.0
lora_alpha: 16
target_modules: ["query", "key", "value", "intermediate.dense", "output.dense"]
modules_to_save: ["classifier"]
Training config
class: AdamW
betas: (0.9, 0.98)
weight_decay: 0.01
learning rate: 1e-4
epoch: 1
batch size: 100
precision: 16-mixed