File size: 970 Bytes
167ed2d 0cbbdb1 167ed2d 0cbbdb1 167ed2d 0cbbdb1 167ed2d 0cbbdb1 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 | ---
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`. |