Buckets:
1.3 GB
12 files
Updated 23 days ago
Ctrl+K
| Name | Size | Uploaded | Xet hash |
|---|---|---|---|
| .gitattributes | 1.52 kB xet | 818ba6de | |
| LICENSE | 19.3 kB xet | 5b936d27 | |
| README.md | 1.37 kB xet | cc3248be | |
| config.json | 1.63 kB xet | de4eb350 | |
| configuration_proteinsage.py | 3.16 kB xet | cad4e2c2 | |
| generation_config.json | 69 Bytes xet | 2055595c | |
| modeling_proteinsage.py | 69.3 kB xet | 3cece0e5 | |
| pytorch_model.bin | 1.3 GB xet | bf4b9855 | |
| special_tokens_map.json | 260 Bytes xet | ab65e328 | |
| tokenization_proteinsage.py | 4.99 kB xet | 2cbac464 | |
| tokenizer.model | 112 Bytes xet | bf227f0b | |
| tokenizer_config.json | 2.04 kB xet | 91b6c34f |
ProteinSage-650
ProteinSage-650M is a protein language model checkpoint for extracting sequence representations from protein sequences.
Usage
import torch
from transformers import AutoModel, AutoTokenizer
model_id = "ProteinSage650"
tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)
model = AutoModel.from_pretrained(
model_id,
trust_remote_code=True,
torch_dtype=torch.bfloat16,
).eval()
seq = "MILMCQHFSGQFSKYFLAVSSDFCHFVFPIILVSHVNFKQMKRKGFALWNDRAVPFTQGIFTTVMILLQYLHGTGM"
inputs = tokenizer(seq, add_special_tokens=True, return_tensors="pt")
with torch.inference_mode():
outputs = model(**inputs, output_hidden_states=True, output_attentions=True, return_dict=True)
hidden_states = outputs.hidden_states
attentions = outputs.attentions
Use trust_remote_code=True because this checkpoint includes custom ProteinSage model and tokenizer code.
Notes
- The checkpoint was verified with
transformersand PyTorch in theproteinsageenvironment. - It is intended primarily for protein sequence representation extraction.
- During loading,
output_layer.weightmay be newly initialized, so masked language modeling predictions may require downstream training or a matching head checkpoint.
License
CC BY-NC 4.0.
- Total size
- 1.3 GB
- Files
- 12
- Last updated
- Oct 26
- Pre-warmed CDN
- US EU US EU