Update README.md
Browse files
README.md
CHANGED
|
@@ -11,9 +11,9 @@ tags:
|
|
| 11 |
- transformers
|
| 12 |
---
|
| 13 |
|
| 14 |
-
# BindScreen Finetuning
|
| 15 |
|
| 16 |
-
This model corresponds to the BinScreen finetuning configuration, in which ESM2 T36 and the projection layers are trained on filtered ChEMBL.
|
| 17 |
|
| 18 |
\[[Github Repo](https://github.com/pcdslab/BindScreen)\] | \[[Dataset on HuggingFace](https://huggingface.co/datasets/SaeedLab/BindScreen)\] | \[[Model Collection](https://huggingface.co/collections/SaeedLab/bindscreen)\] | \[[Cite](#citation)\]
|
| 19 |
|
|
@@ -27,10 +27,12 @@ BindScreen uses a dual-encoder architecture that independently encodes proteins
|
|
| 27 |
|
| 28 |
|
| 29 |
|
| 30 |
-
|
| 31 |
|
| 32 |
-
- [BindScreen-Frozen](https://huggingface.co/SaeedLab/BindScreen-Frozen): only the projection layers are trained, both encoders are frozen.
|
| 33 |
-
- [BindScreen-Finetuning](https://huggingface.co/SaeedLab/BindScreen-Finetuning): the projection layers and ESM2 T36 are trained, MolDeBERTa MLC is frozen.
|
|
|
|
|
|
|
| 34 |
|
| 35 |
## Usage
|
| 36 |
|
|
@@ -59,7 +61,7 @@ backbone = AutoModel.from_pretrained(
|
|
| 59 |
)
|
| 60 |
backbone = PeftModel.from_pretrained(
|
| 61 |
backbone,
|
| 62 |
-
'SaeedLab/BindScreen-lora'
|
| 63 |
).eval()
|
| 64 |
|
| 65 |
|
|
|
|
| 11 |
- transformers
|
| 12 |
---
|
| 13 |
|
| 14 |
+
# BindScreen Finetuning LIT-PCBA
|
| 15 |
|
| 16 |
+
This model corresponds to the BinScreen finetuning configuration, in which ESM2 T36 and the projection layers are trained on filtered ChEMBL for LIT-PCBA.
|
| 17 |
|
| 18 |
\[[Github Repo](https://github.com/pcdslab/BindScreen)\] | \[[Dataset on HuggingFace](https://huggingface.co/datasets/SaeedLab/BindScreen)\] | \[[Model Collection](https://huggingface.co/collections/SaeedLab/bindscreen)\] | \[[Cite](#citation)\]
|
| 19 |
|
|
|
|
| 27 |
|
| 28 |
|
| 29 |
|
| 30 |
+
Four configurations are available in this collection:
|
| 31 |
|
| 32 |
+
- [BindScreen-Frozen-LIT_PCBA](https://huggingface.co/SaeedLab/BindScreen-Frozen-LIT_PCBA): only the projection layers are trained, both encoders are frozen. LIT-PCBA.
|
| 33 |
+
- [BindScreen-Finetuning-LIT_PCBA](https://huggingface.co/SaeedLab/BindScreen-Finetuning-LIT_PCBA): the projection layers and ESM2 T36 are trained, MolDeBERTa MLC is frozen. LIT-PCBA.
|
| 34 |
+
- [BindScreen-Frozen-DUDE](https://huggingface.co/SaeedLab/BindScreen-Frozen-LIT_PCBA): only the projection layers are trained, both encoders are frozen. DUDE.
|
| 35 |
+
- [BindScreen-Finetuning-DUDE](https://huggingface.co/SaeedLab/BindScreen-Finetuning-LIT_PCBA): the projection layers and ESM2 T36 are trained, MolDeBERTa MLC is frozen. DUDE.
|
| 36 |
|
| 37 |
## Usage
|
| 38 |
|
|
|
|
| 61 |
)
|
| 62 |
backbone = PeftModel.from_pretrained(
|
| 63 |
backbone,
|
| 64 |
+
'SaeedLab/BindScreen-lora-LIT_PCBA'
|
| 65 |
).eval()
|
| 66 |
|
| 67 |
|