Hugging Face's logo

quickmt
/
quickmt-en-vi

Translation
English
Vietnamese
Eval Results
Model card Files Community
quickmt-en-vi
File size: 4,333 Bytes 
f0867da
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
98
99
100
101
102
 
---
language:
- en
- vi
tags:
- translation
license: cc-by-4.0
datasets:
- quickmt/quickmt-train.vi-en
model-index:
- name: quickmt-en-vi
  results:
  - task:
      name: Translation eng-vie
      type: translation
      args: eng-vie
    dataset:
      name: flores101-devtest
      type: flores_101
      args: eng_Latn vie_Latn devtest
    metrics:
    - name: BLEU
      type: bleu
      value: 43.68
    - name: CHRF
      type: chrf
      value: 60.75
    - name: COMET
      type: comet
      value: 87.52
---


# `quickmt-en-vi` Neural Machine Translation Model 

`quickmt-en-vi` is a reasonably fast and reasonably accurate neural machine translation model for translation from `en` into `vi`.


## Model Information

* Trained using [`eole`](https://github.com/eole-nlp/eole)
* 195M parameter transformer 'big' with 8 encoder layers and 2 decoder layers
* 20k sentencepiece vocabularies
* Exported for fast inference to [CTranslate2](https://github.com/OpenNMT/CTranslate2) format
* Training data: https://huggingface.co/datasets/quickmt/quickmt-train.fa-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-vi ./quickmt-en-vi
```

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-vi/", device="auto")

# Translate - set beam size to 5 for higher quality (but slower speed)
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)
```

> 'Tiến sĩ Ehud Ur, giáo sư y khoa tại Đại học Dalhousie ở Halifax, Nova Scotia và chủ tịch bộ phận lâm sàng và khoa học của Hiệp hội Tiểu đường Canada cảnh báo rằng nghiên cứu vẫn còn trong những ngày đầu.'
```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)
```

> 'Tiến sĩ Ehud Ur, Giáo sư Y tại Đại học Dalhousie ở Halifax, Nova Scotia và là chủ tịch bộ phận lâm sàng và khoa học của Hiệp hội Tiểu đường Canada đã cảnh báo rằng nghiên cứu mới này vẫn còn trong những ngày đầu.'


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) ("eng_Latn"->"vie_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 with a larger batch size).

|                                  |   bleu |   chrf2 |   comet22 |   Time (s) |
|:---------------------------------|-------:|--------:|----------:|-----------:|
| quickmt/quickmt-en-vi            |  43.68 |   60.75 |     87.52 |       1.5  |
| Helsinki-NLP/opus-mt-en-vi       |  27.57 |   46.45 |     76.22 |       4.67 |
| facebook/nllb-200-distilled-600M |  39.06 |   57.27 |     86.95 |      24.12 |
| facebook/nllb-200-distilled-1.3B |  41.27 |   59.04 |     88.11 |      42.3  |
| facebook/m2m100_418M             |  33.95 |   52.82 |     82.53 |      20.32 |
| facebook/m2m100_1.2B             |  39.46 |   57.45 |     86.56 |      38.65 |