gsarti commited on
Commit
a2b1177
1 Parent(s): f0a24fe

Initial commit

Browse files
README.md ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - en
4
+ - pl
5
+
6
+ tags:
7
+ - translation
8
+
9
+ license: apache-2.0
10
+ ---
11
+
12
+ ### OPUS Tatoeba English-Polish
13
+
14
+ *This model was obtained by running the script [convert_marian_to_pytorch.py](https://github.com/huggingface/transformers/blob/master/src/transformers/models/marian/convert_marian_to_pytorch.py) with the flag `-m eng-pol`. The original models were trained by [Jörg Tiedemann](https://blogs.helsinki.fi/tiedeman/) using the [MarianNMT](https://marian-nmt.github.io/) library. See all available `MarianMTModel` models on the profile of the [Helsinki NLP](https://huggingface.co/Helsinki-NLP) group.
15
+
16
+ * source language name: English
17
+ * target language name: Polish
18
+ * OPUS readme: [README.md](https://object.pouta.csc.fi/Tatoeba-MT-models/eng-pol/README.md)
19
+
20
+ * model: transformer
21
+ * source language code: en
22
+ * target language code: pl
23
+ * dataset: opus
24
+ * release date: 2021-02-19
25
+ * pre-processing: normalization + SentencePiece (spm32k,spm32k)
26
+ * download original weights: [opus-2021-02-19.zip](https://object.pouta.csc.fi/Tatoeba-MT-models/eng-pol/opus-2021-02-19.zip/eng-pol/opus-2021-02-19.zip)
27
+ * Training data:
28
+ * eng-pol: Tatoeba-train (59742979)
29
+ * Validation data:
30
+ * eng-pol: Tatoeba-dev, 44146
31
+ * total-size-shuffled: 44145
32
+ * devset-selected: top 5000 lines of Tatoeba-dev.src.shuffled!
33
+ * Test data:
34
+ * Tatoeba-test.eng-pol: 10000/64925
35
+ * test set translations file: [test.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/eng-pol/opus-2021-02-19.zip/eng-pol/opus-2021-02-19.test.txt)
36
+ * test set scores file: [eval.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/eng-pol/opus-2021-02-19.zip/eng-pol/opus-2021-02-19.eval.txt)
37
+ * BLEU-scores
38
+ |Test set|score|
39
+ |---|---|
40
+ |Tatoeba-test.eng-pol|47.5|
41
+ * chr-F-scores
42
+ |Test set|score|
43
+ |---|---|
44
+ |Tatoeba-test.eng-pol|0.673|
45
+
46
+ ### System Info:
47
+ * hf_name: eng-pol
48
+ * source_languages: en
49
+ * target_languages: pl
50
+ * opus_readme_url: https://object.pouta.csc.fi/Tatoeba-MT-models/eng-pol/opus-2021-02-19.zip/README.md
51
+ * original_repo: Tatoeba-Challenge
52
+ * tags: ['translation']
53
+ * languages: ['en', 'pl']
54
+ * src_constituents: ['eng']
55
+ * tgt_constituents: ['pol']
56
+ * src_multilingual: False
57
+ * tgt_multilingual: False
58
+ * helsinki_git_sha: 70b0a9621f054ef1d8ea81f7d55595d7f64d19ff
59
+ * transformers_git_sha: 7c6cd0ac28f1b760ccb4d6e4761f13185d05d90b
60
+ * port_machine: databox
61
+ * port_time: 2021-10-18-15:11
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": "eng-pol", "source_languages": "en", "target_languages": "pl", "opus_readme_url": "https://object.pouta.csc.fi/Tatoeba-MT-models/eng-pol/opus-2021-02-19.zip/README.md", "original_repo": "Tatoeba-Challenge", "tags": ["translation"], "languages": ["en", "pl"], "src_constituents": ["eng"], "tgt_constituents": ["pol"], "src_multilingual": false, "tgt_multilingual": false, "helsinki_git_sha": "70b0a9621f054ef1d8ea81f7d55595d7f64d19ff", "transformers_git_sha": "7c6cd0ac28f1b760ccb4d6e4761f13185d05d90b", "port_machine": "databox", "port_time": "2021-10-18-15:11"}
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0b4c565a82a950902d1ef3c16c594c412e242d523ea6d8532036b422689afae8
3
+ size 221610115
source.spm ADDED
Binary file (798 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 (834 kB). View file
 
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"source_lang": "eng", "target_lang": "pol", "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-pol", "tokenizer_class": "MarianTokenizer"}
vocab.json ADDED
The diff for this file is too large to render. See raw diff