|
--- |
|
language: |
|
- en |
|
- de |
|
- es |
|
- ru |
|
- zh |
|
base_model: |
|
- microsoft/mdeberta-v3-base |
|
- Unbabel/XCOMET-XXL |
|
--- |
|
|
|
# XCOMET-lite |
|
|
|
**Links:** [EMNLP 2024](https://aclanthology.org/2024.emnlp-main.1223/) | [Arxiv](https://arxiv.org/abs/2406.14553) | [Github repository](https://github.com/NL2G/xCOMET-lite) |
|
|
|
`XCOMET-lite` is a distilled version of [`Unbabel/XCOMET-XXL`](https://huggingface.co/Unbabel/XCOMET-XXL) — a machine translation evaluation model trained to provide an overall quality score between 0 and 1, where 1 represents a perfect translation. |
|
|
|
This model uses [`microsoft/mdeberta-v3-base`](https://huggingface.co/microsoft/deberta-v3-base) as its backbone and has 278 million parameters, making it approximately 38 times smaller than the 10.7 billion-parameter `XCOMET-XXL`. |
|
|
|
## Quick Start |
|
|
|
1. Clone the [GitHub repository](https://github.com/NL2G/xCOMET-lite). |
|
2. Create a conda environment as instructed in the README. |
|
|
|
Then, run the following code: |
|
|
|
``` |
|
from xcomet.deberta_encoder import XCOMETLite |
|
|
|
model = XCOMETLite().from_pretrained("myyycroft/XCOMET-lite") |
|
data = [ |
|
{ |
|
"src": "Elon Musk has acquired Twitter and plans significant changes.", |
|
"mt": "Илон Маск приобрел Twitter и планировал значительные искажения.", |
|
"ref": "Илон Маск приобрел Twitter и планирует значительные изменения." |
|
}, |
|
{ |
|
"src": "Elon Musk has acquired Twitter and plans significant changes.", |
|
"mt": "Илон Маск приобрел Twitter.", |
|
"ref": "Илон Маск приобрел Twitter и планирует значительные изменения." |
|
} |
|
] |
|
|
|
model_output = model.predict(data, batch_size=2, gpus=1) |
|
|
|
print("Segment-level scores:", model_output.scores) |
|
``` |