Initial commit
Browse files- README.md +67 -0
- config.json +43 -0
- metadata.json +1 -0
- pytorch_model.bin +3 -0
- source.spm +0 -0
- special_tokens_map.json +1 -0
- target.spm +0 -0
- tokenizer_config.json +1 -0
- vocab.json +0 -0
README.md
ADDED
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- en
|
4 |
+
- ro
|
5 |
+
|
6 |
+
tags:
|
7 |
+
- translation
|
8 |
+
|
9 |
+
license: apache-2.0
|
10 |
+
---
|
11 |
+
### en-ro
|
12 |
+
|
13 |
+
* source group: English
|
14 |
+
* target group: Romanian
|
15 |
+
* OPUS readme: [eng-ron](https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/eng-ron/README.md)
|
16 |
+
|
17 |
+
* model: transformer-align
|
18 |
+
* source language(s): eng
|
19 |
+
* target language(s): mol ron
|
20 |
+
* model: transformer-align
|
21 |
+
* pre-processing: normalization + SentencePiece (spm32k,spm32k)
|
22 |
+
* a sentence initial language token is required in the form of `>>id<<` (id = valid target language ID)
|
23 |
+
* valid language labels:
|
24 |
+
* download original weights: [opus+bt-2021-03-07.zip](https://object.pouta.csc.fi/Tatoeba-MT-models/eng-ron/opus+bt-2021-03-07.zip)
|
25 |
+
* test set translations: [opus+bt-2021-03-07.test.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/eng-ron/opus+bt-2021-03-07.test.txt)
|
26 |
+
* test set scores: [opus+bt-2021-03-07.eval.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/eng-ron/opus+bt-2021-03-07.eval.txt)
|
27 |
+
|
28 |
+
## Benchmarks
|
29 |
+
|
30 |
+
| testset | BLEU | chr-F | #sent | #words | BP |
|
31 |
+
|---------|-------|-------|-------|--------|----|
|
32 |
+
| newsdev2016-enro.eng-ron | 33.5 | 0.610 | 1999 | 51566 | 0.984 |
|
33 |
+
| newstest2016-enro.eng-ron | 31.7 | 0.591 | 1999 | 49094 | 0.998 |
|
34 |
+
| Tatoeba-test.eng-ron | 46.9 | 0.678 | 5000 | 36851 | 0.983 |
|
35 |
+
|
36 |
+
|
37 |
+
### System Info:
|
38 |
+
- hf_name: en-ro
|
39 |
+
- source_languages: eng
|
40 |
+
- target_languages: ron
|
41 |
+
- opus_readme_url: https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/eng-ron/README.md
|
42 |
+
- original_repo: Tatoeba-Challenge
|
43 |
+
- tags: ['translation']
|
44 |
+
- languages: ['en', 'ro']
|
45 |
+
- src_constituents: ('English', {'eng'})
|
46 |
+
- tgt_constituents: ('Romanian', {'ron'})
|
47 |
+
- src_multilingual: False
|
48 |
+
- tgt_multilingual: False
|
49 |
+
- long_pair: eng-ron
|
50 |
+
- prepro: normalization + SentencePiece (spm32k,spm32k)
|
51 |
+
- url_model: https://object.pouta.csc.fi/Tatoeba-MT-models/eng-ron/opus+bt-2021-03-07.zip
|
52 |
+
- url_test_set: https://object.pouta.csc.fi/Tatoeba-MT-models/eng-ron/opus+bt-2021-03-07.test.txt
|
53 |
+
- src_alpha3: eng
|
54 |
+
- tgt_alpha3: ron
|
55 |
+
- chrF2_score: 0.678
|
56 |
+
- bleu: 46.9
|
57 |
+
- src_name: English
|
58 |
+
- tgt_name: Romanian
|
59 |
+
- train_date: 2021-03-07 00:00:00
|
60 |
+
- src_alpha2: en
|
61 |
+
- tgt_alpha2: ro
|
62 |
+
- prefer_old: False
|
63 |
+
- short_pair: en-ro
|
64 |
+
- helsinki_git_sha: 2ef219d5b67f0afb0c6b732cd07001d84181f002
|
65 |
+
- transformers_git_sha: 12b4d66a80419db30a15e7b9d4208ceb9887c03b
|
66 |
+
- port_machine: LM0-400-22516.local
|
67 |
+
- port_time: 2021-11-08-09:31
|
config.json
ADDED
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"activation_dropout": 0.0,
|
3 |
+
"activation_function": "swish",
|
4 |
+
"architectures": [
|
5 |
+
"MarianMTModel"
|
6 |
+
],
|
7 |
+
"attention_dropout": 0.0,
|
8 |
+
"bad_words_ids": [
|
9 |
+
[
|
10 |
+
65000
|
11 |
+
]
|
12 |
+
],
|
13 |
+
"bos_token_id": 0,
|
14 |
+
"classifier_dropout": 0.0,
|
15 |
+
"d_model": 512,
|
16 |
+
"decoder_attention_heads": 8,
|
17 |
+
"decoder_ffn_dim": 2048,
|
18 |
+
"decoder_layerdrop": 0.0,
|
19 |
+
"decoder_layers": 6,
|
20 |
+
"decoder_start_token_id": 65000,
|
21 |
+
"dropout": 0.1,
|
22 |
+
"encoder_attention_heads": 8,
|
23 |
+
"encoder_ffn_dim": 2048,
|
24 |
+
"encoder_layerdrop": 0.0,
|
25 |
+
"encoder_layers": 6,
|
26 |
+
"eos_token_id": 0,
|
27 |
+
"forced_eos_token_id": 0,
|
28 |
+
"init_std": 0.02,
|
29 |
+
"is_encoder_decoder": true,
|
30 |
+
"max_length": 512,
|
31 |
+
"max_position_embeddings": 512,
|
32 |
+
"model_type": "marian",
|
33 |
+
"normalize_embedding": false,
|
34 |
+
"num_beams": 6,
|
35 |
+
"num_hidden_layers": 6,
|
36 |
+
"pad_token_id": 65000,
|
37 |
+
"scale_embedding": true,
|
38 |
+
"static_position_embeddings": true,
|
39 |
+
"torch_dtype": "float16",
|
40 |
+
"transformers_version": "4.11.3",
|
41 |
+
"use_cache": true,
|
42 |
+
"vocab_size": 65001
|
43 |
+
}
|
metadata.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"hf_name":"en-ro","source_languages":"eng","target_languages":"ron","opus_readme_url":"https:\/\/github.com\/Helsinki-NLP\/Tatoeba-Challenge\/tree\/master\/models\/eng-ron\/README.md","original_repo":"Tatoeba-Challenge","tags":["translation"],"languages":["en","ro"],"src_constituents":["English",["eng"]],"tgt_constituents":["Romanian",["ron"]],"src_multilingual":false,"tgt_multilingual":false,"long_pair":"eng-ron","prepro":" normalization + SentencePiece (spm32k,spm32k)","url_model":"https:\/\/object.pouta.csc.fi\/Tatoeba-MT-models\/eng-ron\/opus+bt-2021-03-07.zip","url_test_set":"https:\/\/object.pouta.csc.fi\/Tatoeba-MT-models\/eng-ron\/opus+bt-2021-03-07.test.txt","src_alpha3":"eng","tgt_alpha3":"ron","chrF2_score":0.678,"bleu":46.9,"src_name":"English","tgt_name":"Romanian","train_date":1615075200000,"src_alpha2":"en","tgt_alpha2":"ro","prefer_old":false,"short_pair":"en-ro","helsinki_git_sha":"2ef219d5b67f0afb0c6b732cd07001d84181f002","transformers_git_sha":"12b4d66a80419db30a15e7b9d4208ceb9887c03b","port_machine":"LM0-400-22516.local","port_time":"2021-11-08-09:31"}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:27bef0a4fce2ae1f4bf5408c6c1453e1bc984d1ecdc62e0455488b8fac465987
|
3 |
+
size 221610115
|
source.spm
ADDED
Binary file (792 kB). View file
|
|
special_tokens_map.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>"}
|
target.spm
ADDED
Binary file (817 kB). View file
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"source_lang": "eng", "target_lang": "ron", "unk_token": "<unk>", "eos_token": "</s>", "pad_token": "<pad>", "model_max_length": 512, "sp_model_kwargs": {}, "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "marian_ckpt/eng-ron", "tokenizer_class": "MarianTokenizer"}
|
vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|