Translation
Fairseq
English
Icelandic
wmt
File size: 2,326 Bytes
d7b83b2
 
 
 
 
 
 
 
 
 
 
 
 
 
ec91a4f
d7b83b2
 
 
 
 
 
 
 
 
 
 
 
 
2767f79
 
d7b83b2
80e5d88
d7b83b2
 
 
 
bb674c0
 
 
 
 
 
 
 
 
d7b83b2
 
7b7e531
 
 
 
 
d7b83b2
 
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
---
license: apache-2.0
language:
- en
- is
library_name: fairseq
tags:
- translation
- wmt
---

## Model description
This is a translation model which translates text from English to Icelandic. It follows the architecture of the transformer model described in [Attention is All You Need](https://arxiv.org/pdf/1706.03762) and was trained with [fairseq](https://github.com/facebookresearch/fairseq) for [WMT24](https://www2.statmt.org/wmt24/).

This is the base version of our model. See also: [wmt24-en-is-transformer-base-deep](https://huggingface.co/arnastofnun/wmt24-en-is-transformer-base-deep), [wmt24-en-is-transformer-big](https://huggingface.co/arnastofnun/wmt24-en-is-transformer-big), [wmt24-en-is-transformer-big-deep](https://huggingface.co/arnastofnun/wmt24-en-is-transformer-big-deep).

| model | d_model | d_ff | h | N_enc | N_dec |
|:---------------|:----------------------|:-------------------|:--------------|:--------------------|:--------------------|
| Base | 512 | 2048 | 8 | 6 | 6 |
| Base_deep | 512 | 2048 | 8 | 36 | 12 |
| Big | 1024 | 4096 | 16 | 6 | 6 |
| Big_deep | 1024 | 4096 | 16 | 36 | 12 |


#### How to use

```python
from fairseq.models.transformer import TransformerModel
TRANSLATION_MODEL_NAME = 'checkpoint_best.pt'
TRANSLATION_MODEL = TransformerModel.from_pretrained('path/to/model', checkpoint_file=TRANSLATION_MODEL_NAME, bpe='sentencepiece', sentencepiece_model='sentencepiece.bpe.model')
src_sentences = ['This is a test sentence.', 'This is another test sentence.']
translated_sentences = TRANSLATION_MODEL.translate(src_sentences)
print(translated_sentences)
```

## Eval results
We evaluated our data on the [WMT21 test set](https://github.com/wmt-conference/wmt21-news-systems/). These are the chrF scores for our published models:

| model  | chrF |
|:---------------|:------|
| Base        | 56.8 |
| Base_deep | 57.1 |
| Big         | 57.7 |
| Big_deep  | 57.7 |
## BibTeX entry and citation info

```bibtex
@inproceedings{jasonarson2024cogsinamachine,
    year={2024},
    title={Cogs in a Machine, Doing What They’re Meant to Do \\– The AMI Submission to the WMT24 General Translation Task},
    author={Atli Jasonarson, Hinrik Hafsteinsson, Bjarki Ármannsson, Steinþór Steingrímsson},
    organization={The Árni Magnússon Institute for Icelandic Studies}
}
```