|
|
--- |
|
|
language: |
|
|
- en |
|
|
- he |
|
|
tags: |
|
|
- translation |
|
|
license: cc-by-4.0 |
|
|
datasets: |
|
|
- quickmt/quickmt-train.he-en |
|
|
model-index: |
|
|
- name: quickmt-en-he |
|
|
results: |
|
|
- task: |
|
|
name: Translation eng-heb |
|
|
type: translation |
|
|
args: eng-heb |
|
|
dataset: |
|
|
name: flores101-devtest |
|
|
type: flores_101 |
|
|
args: eng_Latn heb_Hebr devtest |
|
|
metrics: |
|
|
- name: BLEU |
|
|
type: bleu |
|
|
value: 34.32 |
|
|
- name: CHRF |
|
|
type: chrf |
|
|
value: 62.37 |
|
|
- name: COMET |
|
|
type: comet |
|
|
value: 87.91 |
|
|
--- |
|
|
|
|
|
|
|
|
# `quickmt-en-he` Neural Machine Translation Model |
|
|
|
|
|
`quickmt-en-he` is a reasonably fast and reasonably accurate neural machine translation model for translation from `en` into `he`. |
|
|
|
|
|
|
|
|
## Try it on our Huggingface Space |
|
|
|
|
|
Give it a try before downloading here: https://huggingface.co/spaces/quickmt/QuickMT-Demo |
|
|
|
|
|
|
|
|
## Model Information |
|
|
|
|
|
* Trained using [`eole`](https://github.com/eole-nlp/eole) |
|
|
* 200M parameter transformer 'big' with 8 encoder layers and 2 decoder layers |
|
|
* 32k separate Sentencepiece vocabs |
|
|
* Expested for fast inference to [CTranslate2](https://github.com/OpenNMT/CTranslate2) format |
|
|
* Training data: https://huggingface.co/datasets/quickmt/quickmt-train.he-en/tree/main |
|
|
|
|
|
See the `eole` model configuration in this repository for further details and the `eole-model` for the raw `eole` (pytorch) model. |
|
|
|
|
|
|
|
|
## 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-en-he ./quickmt-en-he |
|
|
``` |
|
|
|
|
|
Finally use the model in python: |
|
|
|
|
|
```python |
|
|
from quickmt import Translator |
|
|
|
|
|
# Auto-detects GPU, set to "cpu" to force CPU inference |
|
|
t = Translator("./quickmt-en-he/", device="auto") |
|
|
|
|
|
# Translate - set beam size to 1 for faster speed (but lower quality) |
|
|
sample_text = 'Dr. Ehud Ur, professor of medicine at Dalhousie University in Halifax, Nova Scotia and chair of the clinical and scientific division of the Canadian Diabetes Association cautioned that the research is still in its early days.' |
|
|
|
|
|
t(sample_text, beam_size=5) |
|
|
``` |
|
|
|
|
|
> '讚"专 讗讛讜讚 讗讜专, 驻专讜驻住讜专 诇专驻讜讗讛 讘讗讜谞讬讘专住讬讟转 讚诇讛讜讝讬 讘讛诇讬驻拽住, 谞讜讘讛 住拽讜讟讬讛 讜讬讜"专 讛诪讞诇拽讛 讛拽诇讬谞讬转 讜讛诪讚注讬转 砖诇 讛讗讙讜讚讛 讛拽谞讚讬转 诇住讜讻专转, 讛讝讛讬专 讻讬 讛诪讞拽专 谞诪爪讗 注讚讬讬谉 讘讬诪讬讜 讛专讗砖讜谞讬诐.' |
|
|
|
|
|
```python |
|
|
# 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) |
|
|
``` |
|
|
|
|
|
> '驻专讜驻住讜专 诇专驻讜讗讛 讘讗讜谞讬讘专住讬讟转 讚诇讛讜住专讛 讘讛诇讬驻拽住 讘谞讜讘讛 住拽讜讟讬讛 讜讬讜"专 讛讞讟讬讘讛 讛拽诇讬谞讬转 讜讛诪讞拽专讬转 砖诇 讗讙讜讚转 讛住讜讻专转 讛拽谞讚讬转 讛讝讛讬专 讻讬 讛诪讞拽专 谞诪爪讗 讻讬讜诐 讘转讞讬诇转讜.' |
|
|
|
|
|
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`. A model in safetensors format to be used with `eole` is also provided. |
|
|
|
|
|
|
|
|
## 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) ("eng_Latn"->"heb_Hebr"). `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 the flores-devtest dataset (1012 sentences) on an Nvidia RTX 4070s GPU with batch size 32. |
|
|
|
|
|
|
|
|
| | bleu | chrf2 | comet22 | Time (s) | |
|
|
|:---------------------------------|-------:|--------:|----------:|-----------:| |
|
|
| quickmt/quickmt-en-he | 34.32 | 62.37 | 87.91 | 1.15 | |
|
|
| facebook/nllb-200-distilled-600M | 23.83 | 53.98 | 84.12 | 25.78 | |
|
|
| facebook/nllb-200-distilled-1.3B | 29 | 58.64 | 87.23 | 44.79 | |
|
|
| facebook/m2m100_418M | 20.53 | 50.74 | 81.38 | 21.7 | |
|
|
| facebook/m2m100_1.2B | 23.78 | 53.73 | 83.81 | 41.71 | |
|
|
|
