File size: 3,696 Bytes
d00d112 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 |
---
language:
- en
- fr
tags:
- translation
license: cc-by-4.0
datasets:
- quickmt/quickmt-train.fr-en
model-index:
- name: quickmt-fr-en
results:
- task:
name: Translation fra-eng
type: translation
args: fra-eng
dataset:
name: flores101-devtest
type: flores_101
args: fra_Latn eng_Latn devtest
metrics:
- name: CHRF
type: chrf
value: 66.77
- name: BLEU
type: bleu
value: 42.17
- name: COMET
type: comet
value: 58.10
---
# `quickmt-fr-en` Neural Machine Translation Model
`quickmt-fr-en` is a reasonably fast and reasonably accurate neural machine translation model for translation from `fr` into `en`.
## Model Information
* Trained using [`eole`](https://github.com/eole-nlp/eole)
* 185M parameter transformer 'big' with 8 encoder layers and 2 decoder layers
* 50k joint Sentencepiece vocabulary
* Exported for fast inference to [CTranslate2](https://github.com/OpenNMT/CTranslate2) format
* Training data: https://huggingface.co/datasets/quickmt/quickmt-train.fr-en/tree/main
See the `eole` model configuration in this repository for further details.
## Usage with `quickmt`
You must install the Nvidia cuda toolkit first, if you want to do GPU inference.
Next, install the `quickmt` python library and download the model:
```bash
git clone https://github.com/quickmt/quickmt.git
pip install ./quickmt/
quickmt-model-download quickmt/quickmt-fr-en ./quickmt-fr-en
```
Finally use the model in python:
```python
from quickmt import Translator
# Auto-detects GPU, set to "cpu" to force CPU inference
t = Translator("./quickmt-fr-en/", device="auto")
# Translate - set beam size to 5 for higher quality (but slower speed)
sample_text = "Résigny est une commune française située dans le département de l'Aisne, en région Hauts-de-France. "
t(sample_text, beam_size=1)
# Get alternative translations by sampling
# You can pass any cTranslate2 `translate_batch` arguments
t([sample_text], sampling_temperature=1.2, beam_size=1, sampling_topk=50, sampling_topp=0.9)
```
The model is in `ctranslate2` format, and the tokenizers are `sentencepiece`, so you can use `ctranslate2` directly instead of through `quickmt`. It is also possible to get this model to work with e.g. [LibreTranslate](https://libretranslate.com/) which also uses `ctranslate2` and `sentencepiece`.
## Metrics
`bleu` and `chrf2` are calculated with [sacrebleu](https://github.com/mjpost/sacrebleu) on the [Flores200 `devtest` test set](https://huggingface.co/datasets/facebook/flores) ("fra_Latn"->"eng_Latn"). `comet22` with the [`comet`](https://github.com/Unbabel/COMET) library and the [default model](https://huggingface.co/Unbabel/wmt22-comet-da). "Time (s)" is the time in seconds to translate (using `ctranslate2`) the flores-devtest dataset (1012 sentences) on an RTX 4070s GPU with batch size 32 (faster speed is possible using a large batch size).
| Model | chrf2 | bleu | comet22 | Time (s) |
| -------------------------------- | ----- | ------- | ------- | -------- |
| quickmt/quickmt-fr-en | 68.22 | 44.28 | 88.86 | 1.1 |
| Helsinki-NLP/opus-mt-fr-en | 66.85 | 41.71 | 88.31 | 3.6 |
| facebook/m2m100_418M | 64.39 | 36.49 | 85.87 | 18.0 |
| facebook/m2m100_1.2B | 66.51 | 41.69 | 88.00 | 34.6 |
| facebook/nllb-200-distilled-600M | 67.82 | 44.04 | 88.47 | 21.7 |
| facebook/nllb-200-distilled-1.3B | 69.30 | 46.22 | 89.24 | 37.1 |
`quickmt-fr-en` is the fastest and is higher quality than `opus-mt-fr-en`, `m2m100_418m`, `m2m100_1.2B` and `nllb-200-distilled-600M`.
|