update model card README.md
Browse files
README.md
CHANGED
|
@@ -1,33 +1,61 @@
|
|
| 1 |
---
|
| 2 |
tags:
|
| 3 |
-
-
|
| 4 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
---
|
| 6 |
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
* source languages: ru
|
| 10 |
-
* target languages: en
|
| 11 |
-
* OPUS readme: [ru-en](https://github.com/Helsinki-NLP/OPUS-MT-train/blob/master/models/ru-en/README.md)
|
| 12 |
-
|
| 13 |
-
* dataset: opus
|
| 14 |
-
* model: transformer-align
|
| 15 |
-
* pre-processing: normalization + SentencePiece
|
| 16 |
-
* download original weights: [opus-2020-02-26.zip](https://object.pouta.csc.fi/OPUS-MT-models/ru-en/opus-2020-02-26.zip)
|
| 17 |
-
* test set translations: [opus-2020-02-26.test.txt](https://object.pouta.csc.fi/OPUS-MT-models/ru-en/opus-2020-02-26.test.txt)
|
| 18 |
-
* test set scores: [opus-2020-02-26.eval.txt](https://object.pouta.csc.fi/OPUS-MT-models/ru-en/opus-2020-02-26.eval.txt)
|
| 19 |
-
|
| 20 |
-
## Benchmarks
|
| 21 |
-
|
| 22 |
-
| testset | BLEU | chr-F |
|
| 23 |
-
|-----------------------|-------|-------|
|
| 24 |
-
| newstest2012.ru.en | 34.8 | 0.603 |
|
| 25 |
-
| newstest2013.ru.en | 27.9 | 0.545 |
|
| 26 |
-
| newstest2014-ruen.ru.en | 31.9 | 0.591 |
|
| 27 |
-
| newstest2015-enru.ru.en | 30.4 | 0.568 |
|
| 28 |
-
| newstest2016-enru.ru.en | 30.1 | 0.565 |
|
| 29 |
-
| newstest2017-enru.ru.en | 33.4 | 0.593 |
|
| 30 |
-
| newstest2018-enru.ru.en | 29.6 | 0.565 |
|
| 31 |
-
| newstest2019-ruen.ru.en | 31.4 | 0.576 |
|
| 32 |
-
| Tatoeba.ru.en | 61.1 | 0.736 |
|
| 33 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
tags:
|
| 3 |
+
- generated_from_trainer
|
| 4 |
+
metrics:
|
| 5 |
+
- bleu
|
| 6 |
+
model-index:
|
| 7 |
+
- name: opus-mt-ru-en-finetuned
|
| 8 |
+
results: []
|
| 9 |
---
|
| 10 |
|
| 11 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 12 |
+
should probably proofread and complete it, then remove this comment. -->
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 13 |
|
| 14 |
+
# opus-mt-ru-en-finetuned
|
| 15 |
+
|
| 16 |
+
This model was trained from scratch on the None dataset.
|
| 17 |
+
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 1.2874
|
| 19 |
+
- Bleu: 34.0221
|
| 20 |
+
- Gen Len: 26.0071
|
| 21 |
+
|
| 22 |
+
## Model description
|
| 23 |
+
|
| 24 |
+
More information needed
|
| 25 |
+
|
| 26 |
+
## Intended uses & limitations
|
| 27 |
+
|
| 28 |
+
More information needed
|
| 29 |
+
|
| 30 |
+
## Training and evaluation data
|
| 31 |
+
|
| 32 |
+
More information needed
|
| 33 |
+
|
| 34 |
+
## Training procedure
|
| 35 |
+
|
| 36 |
+
### Training hyperparameters
|
| 37 |
+
|
| 38 |
+
The following hyperparameters were used during training:
|
| 39 |
+
- learning_rate: 0.0001
|
| 40 |
+
- train_batch_size: 85
|
| 41 |
+
- eval_batch_size: 42
|
| 42 |
+
- seed: 42
|
| 43 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 44 |
+
- lr_scheduler_type: linear
|
| 45 |
+
- num_epochs: 3
|
| 46 |
+
|
| 47 |
+
### Training results
|
| 48 |
+
|
| 49 |
+
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|
| 50 |
+
|:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|
|
| 51 |
+
| 1.343 | 1.0 | 20262 | 1.4699 | 31.0248 | 25.9972 |
|
| 52 |
+
| 1.2232 | 2.0 | 40524 | 1.3449 | 32.8246 | 26.1448 |
|
| 53 |
+
| 1.1324 | 3.0 | 60786 | 1.2874 | 34.0221 | 26.0071 |
|
| 54 |
+
|
| 55 |
+
|
| 56 |
+
### Framework versions
|
| 57 |
+
|
| 58 |
+
- Transformers 4.16.2
|
| 59 |
+
- Pytorch 1.10.2+cu113
|
| 60 |
+
- Datasets 1.18.3
|
| 61 |
+
- Tokenizers 0.11.0
|