Update README.md
Browse files
README.md
CHANGED
|
@@ -5,8 +5,10 @@ base_model:
|
|
| 5 |
pipeline_tag: translation
|
| 6 |
---
|
| 7 |
# Heavy2Light
|
| 8 |
-
Heavy2Light is an seq2seq model designed to generate light chain antibody sequences from corresponding heavy chain inputs. It leverages [HeavyBERTa](https://huggingface.co/leaBroe/HeavyBERTa) as the encoder and [LightGPT](https://huggingface.co/leaBroe/LightGPT) as the decoder. The model is fine-tuned on paired antibody chain data from the [OAS](https://opig.stats.ox.ac.uk/webapps/oas/) and [
|
| 9 |
-
For more information, please visit our GitHub [repository](https://github.com/ibmm-unibe-ch/Heavy2Light.git).
|
|
|
|
|
|
|
| 10 |
|
| 11 |
## How to use the model
|
| 12 |
```python
|
|
|
|
| 5 |
pipeline_tag: translation
|
| 6 |
---
|
| 7 |
# Heavy2Light
|
| 8 |
+
Heavy2Light is an seq2seq model designed to generate light chain antibody sequences from corresponding heavy chain inputs. It leverages [HeavyBERTa](https://huggingface.co/leaBroe/HeavyBERTa) as the encoder and [LightGPT](https://huggingface.co/leaBroe/LightGPT) as the decoder. The model is fine-tuned on paired antibody chain data from the [OAS](https://opig.stats.ox.ac.uk/webapps/oas/) and [PLAbDab](https://opig.stats.ox.ac.uk/webapps/plabdab/) databases.
|
| 9 |
+
For more information, please visit our GitHub [repository](https://github.com/ibmm-unibe-ch/Heavy2Light.git).
|
| 10 |
+
|
| 11 |
+
The model utilizes [Adapters](https://github.com/adapter-hub/adapters) for efficient fine-tuning. You can either download the full model weights and adapter from this repository, or directly use the Heavy2Light adapter available in its dedicated [directory](https://huggingface.co/leaBroe/Heavy2Light_adapter) on Hugging Face.
|
| 12 |
|
| 13 |
## How to use the model
|
| 14 |
```python
|