Instructions to use mila-intel/ProtST-esm1b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use mila-intel/ProtST-esm1b with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="mila-intel/ProtST-esm1b", trust_remote_code=True)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("mila-intel/ProtST-esm1b", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
Update modeling_protst.py
Browse files- modeling_protst.py +0 -1
modeling_protst.py
CHANGED
|
@@ -233,7 +233,6 @@ class ProtSTForProteinPropertyPrediction(ProtSTPreTrainedModel):
|
|
| 233 |
|
| 234 |
self.config = config
|
| 235 |
self.protein_model = EsmForProteinRepresentation(config.protein_config)
|
| 236 |
-
self.text_model = BertForPubMed(config.text_config)
|
| 237 |
self.logit_scale = nn.Parameter(torch.ones([]) * math.log(1 / 0.07))
|
| 238 |
self.classifier = ProtSTHead(config.protein_config, out_dim=config.num_labels)
|
| 239 |
|
|
|
|
| 233 |
|
| 234 |
self.config = config
|
| 235 |
self.protein_model = EsmForProteinRepresentation(config.protein_config)
|
|
|
|
| 236 |
self.logit_scale = nn.Parameter(torch.ones([]) * math.log(1 / 0.07))
|
| 237 |
self.classifier = ProtSTHead(config.protein_config, out_dim=config.num_labels)
|
| 238 |
|