Initial commit

README.md

@@ -0,0 +1,65 @@
+---
+language:
+- fr
+- it
+
+tags:
+- translation
+
+license: apache-2.0
+---
+### fr-it
+
+* source group: French 
+* target group: Italian 
+*  OPUS readme: [fra-ita](https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/fra-ita/README.md)
+
+*  model: transformer-align
+* source language(s): fra
+* target language(s): ita
+* raw source language(s): fra
+* raw target language(s): ita
+* model: transformer-align
+* pre-processing: normalization + SentencePiece (spm32k,spm32k)
+* download original weights: [opusTCv20210807-2021-11-11.zip](https://object.pouta.csc.fi/Tatoeba-MT-models/fra-ita/opusTCv20210807-2021-11-11.zip)
+* test set translations: [opusTCv20210807-2021-11-11.test.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/fra-ita/opusTCv20210807-2021-11-11.test.txt)
+* test set scores: [opusTCv20210807-2021-11-11.eval.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/fra-ita/opusTCv20210807-2021-11-11.eval.txt)
+
+## Benchmarks
+
+| testset | BLEU  | chr-F | #sent | #words | BP |
+|---------|-------|-------|-------|--------|----|
+| Tatoeba-test-v2021-08-07.fra-ita 	| 54.8 	| 0.737 	| 10000 	| 61517 	| 0.953 |
+
+
+### System Info: 
+- hf_name: fr-it
+- source_languages: fra
+- target_languages: ita
+- opus_readme_url: https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/fra-ita/README.md
+- original_repo: Tatoeba-Challenge
+- tags: ['translation']
+- languages: ['fr', 'it']
+- src_constituents: ('French', {'fra'})
+- tgt_constituents: ('Italian', {'ita'})
+- src_multilingual: False
+- tgt_multilingual: False
+- long_pair: fra-ita
+- prepro:  normalization + SentencePiece (spm32k,spm32k)
+- url_model: https://object.pouta.csc.fi/Tatoeba-MT-models/fra-ita/opusTCv20210807-2021-11-11.zip
+- url_test_set: https://object.pouta.csc.fi/Tatoeba-MT-models/fra-ita/opusTCv20210807-2021-11-11.test.txt
+- src_alpha3: fra
+- tgt_alpha3: ita
+- chrF2_score: 0.737
+- bleu: 54.8
+- src_name: French
+- tgt_name: Italian
+- train_date: 2021-11-11 00:00:00
+- src_alpha2: fr
+- tgt_alpha2: it
+- prefer_old: False
+- short_pair: fr-it
+- helsinki_git_sha: 7ab0c987850187e0b10342bfc616cd47c027ba18
+- transformers_git_sha: df1f94eb4a18b1a27d27e32040b60a17410d516e
+- port_machine: LM0-400-22516.local
+- port_time: 2021-11-11-16:20

config.json

@@ -0,0 +1,43 @@
+{
+  "activation_dropout": 0.0,
+  "activation_function": "swish",
+  "architectures": [
+    "MarianMTModel"
+  ],
+  "attention_dropout": 0.0,
+  "bad_words_ids": [
+    [
+      56132
+    ]
+  ],
+  "bos_token_id": 0,
+  "classifier_dropout": 0.0,
+  "d_model": 512,
+  "decoder_attention_heads": 8,
+  "decoder_ffn_dim": 2048,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 6,
+  "decoder_start_token_id": 56132,
+  "dropout": 0.1,
+  "encoder_attention_heads": 8,
+  "encoder_ffn_dim": 2048,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 6,
+  "eos_token_id": 0,
+  "forced_eos_token_id": 0,
+  "init_std": 0.02,
+  "is_encoder_decoder": true,
+  "max_length": 512,
+  "max_position_embeddings": 512,
+  "model_type": "marian",
+  "normalize_embedding": false,
+  "num_beams": 6,
+  "num_hidden_layers": 6,
+  "pad_token_id": 56132,
+  "scale_embedding": true,
+  "static_position_embeddings": true,
+  "torch_dtype": "float16",
+  "transformers_version": "4.12.3",
+  "use_cache": true,
+  "vocab_size": 56133
+}

metadata.json

@@ -0,0 +1 @@
+{"hf_name":"fr-it","source_languages":"fra","target_languages":"ita","opus_readme_url":"https:\/\/github.com\/Helsinki-NLP\/Tatoeba-Challenge\/tree\/master\/models\/fra-ita\/README.md","original_repo":"Tatoeba-Challenge","tags":["translation"],"languages":["fr","it"],"src_constituents":["French",["fra"]],"tgt_constituents":["Italian",["ita"]],"src_multilingual":false,"tgt_multilingual":false,"long_pair":"fra-ita","prepro":" normalization + SentencePiece (spm32k,spm32k)","url_model":"https:\/\/object.pouta.csc.fi\/Tatoeba-MT-models\/fra-ita\/opusTCv20210807-2021-11-11.zip","url_test_set":"https:\/\/object.pouta.csc.fi\/Tatoeba-MT-models\/fra-ita\/opusTCv20210807-2021-11-11.test.txt","src_alpha3":"fra","tgt_alpha3":"ita","chrF2_score":0.737,"bleu":54.8,"src_name":"French","tgt_name":"Italian","train_date":1636588800000,"src_alpha2":"fr","tgt_alpha2":"it","prefer_old":false,"short_pair":"fr-it","helsinki_git_sha":"7ab0c987850187e0b10342bfc616cd47c027ba18","transformers_git_sha":"df1f94eb4a18b1a27d27e32040b60a17410d516e","port_machine":"LM0-400-22516.local","port_time":"2021-11-11-16:20"}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4df05e8d32bba3091ffa2b79d66b49c575262def30a4d89bdf5dbecff73dd22f
+size 203430723

special_tokens_map.json

@@ -0,0 +1 @@
+{"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>"}

tokenizer_config.json

@@ -0,0 +1 @@
+{"source_lang": "fra", "target_lang": "ita", "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/fra-ita", "tokenizer_class": "MarianTokenizer"}