Initial commit

README.md

@@ -0,0 +1,97 @@
+---
+language:
+- es
+- zh
+
+tags:
+- translation
+
+license: apache-2.0
+---
+### es-zh
+
+* source group: Spanish 
+* target group: Chinese 
+*  OPUS readme: [spa-zho](https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/spa-zho/README.md)
+
+*  model: transformer
+* source language(s): spa
+* target language(s): cjy_Hans cjy_Hant cmn cmn_Hans cmn_Hant hsn hsn_Hani lzh nan wuu yue_Hans yue_Hant
+* model: transformer
+* pre-processing: normalization + SentencePiece (spm32k,spm32k)
+* a sentence initial language token is required in the form of `>>id<<` (id = valid target language ID)
+* download original weights: [opus-2021-01-04.zip](https://object.pouta.csc.fi/Tatoeba-MT-models/spa-zho/opus-2021-01-04.zip)
+* test set translations: [opus-2021-01-04.test.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/spa-zho/opus-2021-01-04.test.txt)
+* test set scores: [opus-2021-01-04.eval.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/spa-zho/opus-2021-01-04.eval.txt)
+
+## Benchmarks
+
+| testset               | BLEU  | chr-F |
+|-----------------------|-------|-------|
+| Tatoeba-test.spa.zho 	| 38.8 	| 0.324 |
+
+
+### System Info: 
+- hf_name: es-zh
+
+- source_languages: spa
+
+- target_languages: zho
+
+- opus_readme_url: https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/spa-zho/README.md
+
+- original_repo: Tatoeba-Challenge
+
+- tags: ['translation']
+
+- languages: ['es', 'zh']
+
+- src_constituents: ('Spanish', {'spa'})
+
+- tgt_constituents: ('Chinese', {'wuu_Bopo', 'wuu', 'cmn_Hang', 'lzh_Kana', 'lzh', 'wuu_Hani', 'lzh_Yiii', 'yue_Hans', 'cmn_Hani', 'cjy_Hans', 'cmn_Hans', 'cmn_Kana', 'zho_Hans', 'zho_Hant', 'yue', 'cmn_Bopo', 'yue_Hang', 'lzh_Hans', 'wuu_Latn', 'yue_Hant', 'hak_Hani', 'lzh_Bopo', 'cmn_Hant', 'lzh_Hani', 'lzh_Hang', 'cmn', 'lzh_Hira', 'yue_Bopo', 'yue_Hani', 'gan', 'zho', 'cmn_Yiii', 'yue_Hira', 'cmn_Latn', 'yue_Kana', 'cjy_Hant', 'cmn_Hira', 'nan_Hani', 'nan'})
+
+- src_multilingual: False
+
+- tgt_multilingual: False
+
+- long_pair: spa-zho
+
+- prepro:  normalization + SentencePiece (spm32k,spm32k)
+
+- url_model: https://object.pouta.csc.fi/Tatoeba-MT-models/spa-zho/opus-2021-01-04.zip
+
+- url_test_set: https://object.pouta.csc.fi/Tatoeba-MT-models/spa-zho/opus-2021-01-04.test.txt
+
+- src_alpha3: spa
+
+- tgt_alpha3: zho
+
+- chrF2_score: 0.324
+
+- bleu: 38.8
+
+- brevity_penalty: 0.878
+
+- ref_len: 22762.0
+
+- src_name: Spanish
+
+- tgt_name: Chinese
+
+- train_date: 2021-01-04 00:00:00
+
+- src_alpha2: es
+
+- tgt_alpha2: zh
+
+- prefer_old: False
+
+- short_pair: es-zh
+
+- helsinki_git_sha: dfdcef114ffb8a8dbb7a3fcf84bde5af50309500
+
+- transformers_git_sha: 1310e1a758edc8e89ec363db76863c771fbeb1de
+
+- port_machine: LM0-400-22516.local
+
+- port_time: 2021-01-04-18:53

config.json

@@ -0,0 +1,56 @@
+{
+  "activation_dropout": 0.0,
+  "activation_function": "swish",
+  "add_bias_logits": false,
+  "add_final_layer_norm": false,
+  "architectures": [
+    "MarianMTModel"
+  ],
+  "attention_dropout": 0.0,
+  "bad_words_ids": [
+    [
+      65000
+    ]
+  ],
+  "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": 65000,
+  "do_blenderbot_90_layernorm": false,
+  "dropout": 0.1,
+  "encoder_attention_heads": 8,
+  "encoder_ffn_dim": 2048,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 6,
+  "eos_token_id": 0,
+  "extra_pos_embeddings": 2,
+  "force_bos_token_to_be_generated": false,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "init_std": 0.02,
+  "is_encoder_decoder": true,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "max_length": 512,
+  "max_position_embeddings": 512,
+  "model_type": "marian",
+  "normalize_before": false,
+  "normalize_embedding": false,
+  "num_beams": 6,
+  "num_hidden_layers": 6,
+  "pad_token_id": 65000,
+  "scale_embedding": true,
+  "static_position_embeddings": true,
+  "use_cache": true,
+  "vocab_size": 65001
+}

metadata.json

@@ -0,0 +1 @@
+{"hf_name":"es-zh","source_languages":"spa","target_languages":"zho","opus_readme_url":"https:\/\/github.com\/Helsinki-NLP\/Tatoeba-Challenge\/tree\/master\/models\/spa-zho\/README.md","original_repo":"Tatoeba-Challenge","tags":["translation"],"languages":["es","zh"],"src_constituents":["Spanish",["spa"]],"tgt_constituents":["Chinese",["wuu_Bopo","wuu","cmn_Hang","lzh_Kana","lzh","wuu_Hani","lzh_Yiii","yue_Hans","cmn_Hani","cjy_Hans","cmn_Hans","cmn_Kana","zho_Hans","zho_Hant","yue","cmn_Bopo","yue_Hang","lzh_Hans","wuu_Latn","yue_Hant","hak_Hani","lzh_Bopo","cmn_Hant","lzh_Hani","lzh_Hang","cmn","lzh_Hira","yue_Bopo","yue_Hani","gan","zho","cmn_Yiii","yue_Hira","cmn_Latn","yue_Kana","cjy_Hant","cmn_Hira","nan_Hani","nan"]],"src_multilingual":false,"tgt_multilingual":false,"long_pair":"spa-zho","prepro":" normalization + SentencePiece (spm32k,spm32k)","url_model":"https:\/\/object.pouta.csc.fi\/Tatoeba-MT-models\/spa-zho\/opus-2021-01-04.zip","url_test_set":"https:\/\/object.pouta.csc.fi\/Tatoeba-MT-models\/spa-zho\/opus-2021-01-04.test.txt","src_alpha3":"spa","tgt_alpha3":"zho","chrF2_score":0.324,"bleu":38.8,"brevity_penalty":0.878,"ref_len":22762.0,"src_name":"Spanish","tgt_name":"Chinese","train_date":1609718400000,"src_alpha2":"es","tgt_alpha2":"zh","prefer_old":false,"short_pair":"es-zh","helsinki_git_sha":"dfdcef114ffb8a8dbb7a3fcf84bde5af50309500","transformers_git_sha":"1310e1a758edc8e89ec363db76863c771fbeb1de","port_machine":"LM0-400-22516.local","port_time":"2021-01-04-18:53"}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:53269d122666ef14b68d3b410ca89000d9c2b3eddd75fd7e38c7a69035df7620
+size 155055643

special_tokens_map.json

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

tokenizer_config.json

@@ -0,0 +1 @@
+{"source_lang": "spa", "target_lang": "zho", "unk_token": "<unk>", "eos_token": "</s>", "pad_token": "<pad>", "model_max_length": 512, "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "marian_ckpt/spa-zho"}