|
|
--- |
|
|
library_name: transformers |
|
|
tags: [] |
|
|
--- |
|
|
|
|
|
## BALM-shuffled |
|
|
BALM-shuffled is an antibody language model that uses a [RoBERTa](https://arxiv.org/abs/1907.11692) architecture and was pre-trained on **randomly shuffled** paired antibody sequences from [Jaffe et al.](https://www.nature.com/articles/s41586-022-05371-z) |
|
|
This was a control model used to evaluate the benefits of natively paired sequences in [our paper](https://doi.org/10.1016/j.patter.2024.100967) published in Patterns. Therefore, **this model should not be used for real use cases**; use [BALM-paired](https://huggingface.co/brineylab/BALM-paired) instead. |
|
|
|
|
|
### Use |
|
|
Load the model and tokenizer as follows: |
|
|
```python |
|
|
from transformers import RobertaTokenizer, RobertaForMaskedLM |
|
|
|
|
|
model = RobertaForMaskedLM.from_pretrained("brineylab/BALM-shuffled") |
|
|
tokenizer = RobertaTokenizer.from_pretrained("brineylab/BALM-shuffled") |
|
|
``` |
|
|
|
|
|
The tokenizer expects sequences formatted as: `HEAVY_CHAIN</s>LIGHT_CHAIN`. |
