Update README.md
Browse files
README.md
CHANGED
|
@@ -15,11 +15,11 @@ base_model: facebook/esm2_t36_3B_UR50D
|
|
| 15 |
library_name: peft
|
| 16 |
---
|
| 17 |
|
| 18 |
-
# BindScreen - ESM2 LoRA Adapter
|
| 19 |
|
| 20 |
-
This repository contains the LoRA adapter weights for the protein encoder used in BindScreen, trained on filtered ChEMBL. BindScreen is a sequence-based virtual screening method built on a dual-encoder contrastive architecture. The adapter fine-tunes [ESM2 T36](https://huggingface.co/facebook/esm2_t36_3B_UR50D) on protein-molecule interaction task.
|
| 21 |
|
| 22 |
-
The projection layers are available separately at [SaeedLab/BindScreen-Finetuning](https://huggingface.co/SaeedLab/BindScreen-Finetuning), which also contains the full model description, architecture diagram, and usage examples.
|
| 23 |
|
| 24 |
\[[Github Repo](https://github.com/pcdslab/BindScreen)\] | \[[Dataset on HuggingFace](https://huggingface.co/datasets/SaeedLab/BindScreen)\] | \[[Model Collection](https://huggingface.co/collections/SaeedLab/bindscreen)\] | \[[Cite](#citation)\]
|
| 25 |
|
|
@@ -29,10 +29,7 @@ Virtual screening aims to identify candidate molecules that bind to a target pro
|
|
| 29 |
|
| 30 |
## Model Details
|
| 31 |
|
| 32 |
-
This adapter corresponds to the **BindScreen-Finetuning** configuration, in which ESM2 T36 is fine-tuned via LoRA alongside the projection layers.
|
| 33 |
-
|
| 34 |
-
- [BindScreen-Frozen](https://huggingface.co/SaeedLab/BindScreen-Frozen): only the projection layers are trained, both encoders are frozen.
|
| 35 |
-
- [BindScreen-Finetuning](https://huggingface.co/SaeedLab/BindScreen-Finetuning): the projection layers and ESM2 T36 are trained via LoRA, MolDeBERTa MLC is frozen.
|
| 36 |
|
| 37 |
| Field | Value |
|
| 38 |
|---|---|
|
|
@@ -46,7 +43,7 @@ This adapter corresponds to the **BindScreen-Finetuning** configuration, in whic
|
|
| 46 |
|
| 47 |
## Usage
|
| 48 |
|
| 49 |
-
This adapter must be used together with [SaeedLab/BindScreen-Finetuning](https://huggingface.co/SaeedLab/BindScreen-Finetuning), which provides the projection layers. The full usage example, including molecule encoding and similarity computation, is available in that repository.
|
| 50 |
|
| 51 |
### Dependencies
|
| 52 |
|
|
|
|
| 15 |
library_name: peft
|
| 16 |
---
|
| 17 |
|
| 18 |
+
# BindScreen - ESM2 LoRA Adapter - LIT-PCBA
|
| 19 |
|
| 20 |
+
This repository contains the LoRA adapter weights for the protein encoder used in BindScreen, trained on filtered ChEMBL for LIT-PCBA. BindScreen is a sequence-based virtual screening method built on a dual-encoder contrastive architecture. The adapter fine-tunes [ESM2 T36](https://huggingface.co/facebook/esm2_t36_3B_UR50D) on protein-molecule interaction task.
|
| 21 |
|
| 22 |
+
The projection layers are available separately at [SaeedLab/BindScreen-Finetuning-LIT_PCBA](https://huggingface.co/SaeedLab/BindScreen-Finetuning-LIT_PCBA), which also contains the full model description, architecture diagram, and usage examples.
|
| 23 |
|
| 24 |
\[[Github Repo](https://github.com/pcdslab/BindScreen)\] | \[[Dataset on HuggingFace](https://huggingface.co/datasets/SaeedLab/BindScreen)\] | \[[Model Collection](https://huggingface.co/collections/SaeedLab/bindscreen)\] | \[[Cite](#citation)\]
|
| 25 |
|
|
|
|
| 29 |
|
| 30 |
## Model Details
|
| 31 |
|
| 32 |
+
This adapter corresponds to the **BindScreen-Finetuning-LIT_PCBA** configuration, in which ESM2 T36 is fine-tuned via LoRA alongside the projection layers.
|
|
|
|
|
|
|
|
|
|
| 33 |
|
| 34 |
| Field | Value |
|
| 35 |
|---|---|
|
|
|
|
| 43 |
|
| 44 |
## Usage
|
| 45 |
|
| 46 |
+
This adapter must be used together with [SaeedLab/BindScreen-Finetuning-LIT_PCBA](https://huggingface.co/SaeedLab/BindScreen-Finetuning-LIT_PCBA), which provides the projection layers. The full usage example, including molecule encoding and similarity computation, is available in that repository.
|
| 47 |
|
| 48 |
### Dependencies
|
| 49 |
|