README file updated
Browse files
README.md
CHANGED
|
@@ -1,3 +1,69 @@
|
|
| 1 |
---
|
| 2 |
license: mit
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
+
language:
|
| 4 |
+
- en
|
| 5 |
+
- eu
|
| 6 |
+
metrics:
|
| 7 |
+
- BLEU
|
| 8 |
+
- TER
|
| 9 |
+
tags:
|
| 10 |
+
- text2text-generation
|
| 11 |
+
- open-nmt
|
| 12 |
+
- pytorch
|
| 13 |
---
|
| 14 |
+
|
| 15 |
+
# Itzune v1.9 EN -> EU machine translation argos model
|
| 16 |
+
|
| 17 |
+
This model was trained using [argostrain](https://github.com/argosopentech/argos-train) training scripts with 11,542,706 English to Basque parallel strings extracted from datasets obtained directly from the [Opus project](https://opus.nlpl.eu/).
|
| 18 |
+
|
| 19 |
+
## Model description
|
| 20 |
+
|
| 21 |
+
|
| 22 |
+
- **Developed by:** argostranslate
|
| 23 |
+
- **Model type:** traslation
|
| 24 |
+
- **Model version:** v1.9
|
| 25 |
+
- **Source Language:** English
|
| 26 |
+
- **Target Language:** Basque
|
| 27 |
+
- **License:** MIT
|
| 28 |
+
|
| 29 |
+
## Training Data
|
| 30 |
+
|
| 31 |
+
The English-Basque parallel sentences were collected from the following datasets:
|
| 32 |
+
|
| 33 |
+
| Dataset | Sentences before cleaning |
|
| 34 |
+
|----------------------|--------------------------:|
|
| 35 |
+
| CCMatrix v1 | 7,788,871 |
|
| 36 |
+
| OpenSubtitles v2018 | 805,780 |
|
| 37 |
+
| XLEnt v1.2 | 800,631 |
|
| 38 |
+
| GNOME v1 | 652,298 |
|
| 39 |
+
| HPLT v1.1 | 610,694 |
|
| 40 |
+
| EhuHac v1 | 585,210 |
|
| 41 |
+
| WikiMatrix v1 | 119,480 |
|
| 42 |
+
| KDE4 v2 | 100,160 |
|
| 43 |
+
| wikimedia v20230407 | 60,990 |
|
| 44 |
+
| bible-uedin v1 | 15,893 |
|
| 45 |
+
| Tatoeba v2023-04-12 | 2,070 |
|
| 46 |
+
| Wiktionary | 629 |
|
| 47 |
+
| **Total** | **11,542,706** |
|
| 48 |
+
|
| 49 |
+
### Evaluation results
|
| 50 |
+
Below are the evaluation results on the machine translation from English to Basque compared to [Google Translate](https://translate.google.com/), [NLLB 200 3.3B](https://huggingface.co/facebook/nllb-200-3.3B) and [mt-hitz-en-eu](https://huggingface.co/HiTZ/mt-hitz-en-eu):
|
| 51 |
+
|
| 52 |
+
#### BLEU scores
|
| 53 |
+
|
| 54 |
+
| Test set |Google Translate | NLLB 3.3 | mt-hitz-en-eu | itzune 1.9 |
|
| 55 |
+
|----------------------|-----------------|----------|---------------|------------|
|
| 56 |
+
| Flores 200 devtest | **20.5** | 13.3 | 19.2 | 17.0 |
|
| 57 |
+
| TaCON | **12.1** | 9.4 | 8.8 | - |
|
| 58 |
+
| NTREX | **15.7** | 8.0 | 14.5 | - |
|
| 59 |
+
| Average | **16.1** | 10.2 | 14.2 | - |
|
| 60 |
+
|
| 61 |
+
#### TER scores
|
| 62 |
+
|
| 63 |
+
| Test set |Google Translate | NLLB 3.3 | mt-hitz-en-eu | itzune 1.9 |
|
| 64 |
+
|----------------------|-----------------|----------|---------------|------------|
|
| 65 |
+
| Flores 200 devtest |**59.5** | 70.4 | 65.0 | 70.1 |
|
| 66 |
+
| TaCON |**69.5** | 75.3 | 76.8 | - |
|
| 67 |
+
| NTREX |**65.8** | 81.6 | 66.7 | - |
|
| 68 |
+
| Average |**64.9** | 75.8 | 68.2 | - |
|
| 69 |
+
|