|
|
--- |
|
|
language: |
|
|
- en |
|
|
- he |
|
|
tags: |
|
|
- translation |
|
|
license: cc-by-4.0 |
|
|
datasets: |
|
|
- quickmt/quickmt-train.he-en |
|
|
model-index: |
|
|
- name: quickmt-he-en |
|
|
results: |
|
|
- task: |
|
|
name: Translation heb-eng |
|
|
type: translation |
|
|
args: heb-eng |
|
|
dataset: |
|
|
name: flores101-devtest |
|
|
type: flores_101 |
|
|
args: heb_Habr eng_Latn devtest |
|
|
metrics: |
|
|
- name: BLEU |
|
|
type: bleu |
|
|
value: 45.01 |
|
|
- name: CHRF |
|
|
type: chrf |
|
|
value: 68.39 |
|
|
- name: COMET |
|
|
type: comet |
|
|
value: 88.31 |
|
|
--- |
|
|
|
|
|
|
|
|
# `quickmt-he-en` Neural Machine Translation Model |
|
|
|
|
|
`quickmt-he-en` is a reasonably fast and reasonably accurate neural machine translation model for translation from `he` into `en`. |
|
|
|
|
|
|
|
|
## 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 |
|
|
* Exported 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-he-en ./quickmt-he-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-he-en/", device="auto") |
|
|
|
|
|
# Translate - set beam size to 1 for faster speed (but lower quality) |
|
|
sample_text = '"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 讚诇讛讗讜讝讬 讘讛诇讬驻拽住, 谞讜讘讛 住拽讜讟讬讛 讜专讗砖 讛诪讞诇拽讛 讛拽诇讬谞讬转 讜讛诪讚注讬转 砖诇 讗专讙讜谉 讞拽专 讛住讜讻专转 讛拽谞讚讬 讛讝讛讬专 砖讛诪讞拽专 注讚讬讬谉 讘讬诪讬讜 讛专讗砖讜谞讬诐."' |
|
|
|
|
|
t(sample_text, beam_size=5) |
|
|
``` |
|
|
|
|
|
> '"Dr. Ehud Orr, professor of medicine at Dalhousie University in Halifax, Nova Scotia and head of the clinical and scientific department of the Canadian Diabetes Research Organization warned that the study was still in its early days."' |
|
|
|
|
|
```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) |
|
|
``` |
|
|
|
|
|
> '"Deutschland Institute Professor in Medicine at Dalhousie University in Halifax, Nova Scotia, head of Clinical and Scientific Department of the Canadian Diabetes Research Organisation has warned that research was in its early days."' |
|
|
|
|
|
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) ("heb_Hebr"->"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 the flores-devtest dataset (1012 sentences) on an RTX 4070s GPU with batch size 32. |
|
|
|
|
|
| | bleu | chrf2 | comet22 | Time (s) | |
|
|
|:----------------------------------|-------:|--------:|----------:|-----------:| |
|
|
| quickmt/quickmt-he-en | 45.01 | 68.39 | 88.31 | 1.19 | |
|
|
| Helsinki-NLP/opus-mt-tc-big-he-en | 44.04 | 68.21 | 87.87 | 3.28 | |
|
|
| facebook/nllb-200-distilled-600M | 39.71 | 64.1 | 85.75 | 21.43 | |
|
|
| facebook/nllb-200-distilled-1.3B | 44.11 | 67.3 | 87.95 | 37.23 | |
|
|
| facebook/m2m100_418M | 32.2 | 59.16 | 82.44 | 17.92 | |
|
|
| facebook/m2m100_1.2B | 37.36 | 62.68 | 84.92 | 34.94 | |
|
|
|
|
|
|
|
|
|
|
|
|
