README.md

---
language:
- en
- es
tags:
- translation
license: cc-by-4.0
datasets:
- quickmt/quickmt-train.es-en
model-index:
- name: quickmt-es-en
  results:
  - task:
      name: Translation spa-eng
      type: translation
      args: spa-eng
    dataset:
      name: flores101-devtest
      type: flores_101
      args: spa_Latn eng_Latn devtest
    metrics:
    - name: BLEU
      type: bleu
      value: 28.64
    - name: CHRF
      type: chrf
      value: 58.61
    - name: COMET
      type: comet
      value: 86.11
---


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

`quickmt-es-en` is a reasonably fast and reasonably accurate neural machine translation model for translation from `es` 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.it-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](github.com/quickmt/quickmt). 

```bash
git clone https://github.com/quickmt/quickmt.git
pip install ./quickmt/
```

Finally, use the model in python:

```python
from quickmt import Translator
from huggingface_hub import snapshot_download

# Download Model (if not downloaded already) and return path to local model
# Device is either 'auto', 'cpu' or 'cuda'
t = Translator(
    snapshot_download("quickmt/quickmt-es-en", ignore_patterns="eole-model/*"),
    device="cpu"
)

# Translate - set beam size to 1 for faster speed (but lower quality)
sample_text = 'La investigación todavía se ubica en su etapa inicial, conforme indicara el Dr. Ehud Ur, docente en la carrera de medicina de la Universidad de Dalhousie, en Halifax, Nueva Escocia, y director del departamento clínico y científico de la Asociación Canadiense de Diabetes.'
t(sample_text, beam_size=5)

> 'The research is still in its early stages, as indicated by Dr. Ehud Ur, a medical professor at the University of Dalhousie, Halifax, Nova Scotia, and director of the clinical and scientific department of the Canadian Diabetes Association.'

# 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 research is still in its initial stages as instructed by Dr. Ehud Ur, a professor at the medical degree, University of Dalhousie, Halifax, Nova Scotia, and director of the clinical and scientific department of the Canadian Diabetes Association.'
```

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) ("spa_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 the flores-devtest dataset (1012 sentences) on an RTX 4070s GPU with batch size 32 (faster speed is possible using a larger batch size).

|                                  |   bleu |   chrf2 |   comet22 |   Time (s) |
|:---------------------------------|-------:|--------:|----------:|-----------:|
| quickmt/quickmt-es-en            |  28.64 |   58.61 |     86.11 |       1.33 |
| Helsink-NLP/opus-mt-es-en        |  27.62 |   58.38 |     86.01 |       3.67 |
| facebook/nllb-200-distilled-600M |  30.02 |   59.71 |     86.55 |      21.99 |
| facebook/nllb-200-distilled-1.3B |  31.58 |   60.96 |     87.25 |      38.2  |
| facebook/m2m100_418M             |  22.85 |   55.04 |     82.9  |      18.83 |
| facebook/m2m100_1.2B             |  26.84 |   57.69 |     85.47 |      36.22 |