Transformers
Safetensors
English
esmfold2
biology
esm
protein
protein-structure-prediction
structure-prediction
protein-design
3d-structure
confidence-estimation
molecular-dynamics
Instructions to use biohub/ESMFold2-Fast with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use biohub/ESMFold2-Fast with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("biohub/ESMFold2-Fast", dtype="auto") - Notebooks
- Google Colab
- Kaggle
Upload folder using huggingface_hub
Browse files- config.json +3 -3
config.json
CHANGED
|
@@ -20,7 +20,7 @@
|
|
| 20 |
"d_single": 384,
|
| 21 |
"disable_msa_features": false,
|
| 22 |
"dtype": "float32",
|
| 23 |
-
"esmc_id": "biohub/
|
| 24 |
"folding_trunk": {
|
| 25 |
"dropout": 0.25,
|
| 26 |
"n_heads": 8,
|
|
@@ -48,7 +48,7 @@
|
|
| 48 |
"enabled": true,
|
| 49 |
"lm_dropout": 0.25,
|
| 50 |
"n_layers": 4,
|
| 51 |
-
"
|
| 52 |
},
|
| 53 |
"lm_num_layers": 80,
|
| 54 |
"model_type": "esmfold2_v2",
|
|
@@ -64,7 +64,7 @@
|
|
| 64 |
"n_relative_chain_bins": 2,
|
| 65 |
"n_relative_residx_bins": 32,
|
| 66 |
"num_diffusion_samples": 32,
|
| 67 |
-
"
|
| 68 |
"parcae": {
|
| 69 |
"coda_n_layers": 2,
|
| 70 |
"enabled": true,
|
|
|
|
| 20 |
"d_single": 384,
|
| 21 |
"disable_msa_features": false,
|
| 22 |
"dtype": "float32",
|
| 23 |
+
"esmc_id": "biohub/ESMC-6B",
|
| 24 |
"folding_trunk": {
|
| 25 |
"dropout": 0.25,
|
| 26 |
"n_heads": 8,
|
|
|
|
| 48 |
"enabled": true,
|
| 49 |
"lm_dropout": 0.25,
|
| 50 |
"n_layers": 4,
|
| 51 |
+
"per_loop_lm_dropout": true
|
| 52 |
},
|
| 53 |
"lm_num_layers": 80,
|
| 54 |
"model_type": "esmfold2_v2",
|
|
|
|
| 64 |
"n_relative_chain_bins": 2,
|
| 65 |
"n_relative_residx_bins": 32,
|
| 66 |
"num_diffusion_samples": 32,
|
| 67 |
+
"num_loops": 3,
|
| 68 |
"parcae": {
|
| 69 |
"coda_n_layers": 2,
|
| 70 |
"enabled": true,
|