Initial commit

README.md

@@ -1,24 +1,96 @@
 ---
+language:
+- es
+- he
+
 tags:
 - translation
----
 
-### opus-mt-es-he
+license: apache-2.0
+---
+### es-he
 
-* source languages: es
-* target languages: he
-*  OPUS readme: [es-he](https://github.com/Helsinki-NLP/OPUS-MT-train/blob/master/models/es-he/README.md)
+* source group: Spanish 
+* target group: Hebrew 
+*  OPUS readme: [spa-heb](https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/spa-heb/README.md)
 
-*  dataset: opus
-* model: transformer-align
-* pre-processing: normalization + SentencePiece
-* download original weights: [opus-2020-01-16.zip](https://object.pouta.csc.fi/OPUS-MT-models/es-he/opus-2020-01-16.zip)
-* test set translations: [opus-2020-01-16.test.txt](https://object.pouta.csc.fi/OPUS-MT-models/es-he/opus-2020-01-16.test.txt)
-* test set scores: [opus-2020-01-16.eval.txt](https://object.pouta.csc.fi/OPUS-MT-models/es-he/opus-2020-01-16.eval.txt)
+*  model: transformer
+* source language(s): spa
+* target language(s): heb
+* model: transformer
+* pre-processing: normalization + SentencePiece (spm32k,spm32k)
+* download original weights: [opus-2020-12-10.zip](https://object.pouta.csc.fi/Tatoeba-MT-models/spa-heb/opus-2020-12-10.zip)
+* test set translations: [opus-2020-12-10.test.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/spa-heb/opus-2020-12-10.test.txt)
+* test set scores: [opus-2020-12-10.eval.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/spa-heb/opus-2020-12-10.eval.txt)
 
 ## Benchmarks
 
 | testset               | BLEU  | chr-F |
 |-----------------------|-------|-------|
-| Tatoeba.es.he 	| 44.3 	| 0.642 |
+| Tatoeba-test.spa.heb 	| 43.6 	| 0.636 |
+
+
+### System Info: 
+- hf_name: es-he
+
+- source_languages: spa
+
+- target_languages: heb
+
+- opus_readme_url: https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/spa-heb/README.md
+
+- original_repo: Tatoeba-Challenge
+
+- tags: ['translation']
+
+- languages: ['es', 'he']
+
+- src_constituents: ('Spanish', {'spa'})
+
+- tgt_constituents: ('Hebrew', {'heb'})
+
+- src_multilingual: False
+
+- tgt_multilingual: False
+
+- long_pair: spa-heb
+
+- prepro:  normalization + SentencePiece (spm32k,spm32k)
+
+- url_model: https://object.pouta.csc.fi/Tatoeba-MT-models/spa-heb/opus-2020-12-10.zip
+
+- url_test_set: https://object.pouta.csc.fi/Tatoeba-MT-models/spa-heb/opus-2020-12-10.test.txt
+
+- src_alpha3: spa
+
+- tgt_alpha3: heb
+
+- chrF2_score: 0.636
+
+- bleu: 43.6
+
+- brevity_penalty: 0.992
+
+- ref_len: 12112.0
+
+- src_name: Spanish
+
+- tgt_name: Hebrew
+
+- train_date: 2020-12-10 00:00:00
+
+- src_alpha2: es
+
+- tgt_alpha2: he
+
+- prefer_old: False
+
+- short_pair: es-he
+
+- helsinki_git_sha: b317f78a3ec8a556a481b6a53dc70dc11769ca96
+
+- transformers_git_sha: 1310e1a758edc8e89ec363db76863c771fbeb1de
+
+- port_machine: LM0-400-22516.local
 
+- port_time: 2020-12-11-11:41

config.json

@@ -1,5 +1,4 @@
 {
-  "_num_labels": 3,
   "activation_dropout": 0.0,
   "activation_function": "swish",
   "add_bias_logits": false,
@@ -10,23 +9,26 @@
   "attention_dropout": 0.0,
   "bad_words_ids": [
     [
-      64683
+      63104
     ]
   ],
   "bos_token_id": 0,
-  "classif_dropout": 0.0,
+  "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": 64683,
+  "decoder_start_token_id": 63104,
+  "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",
@@ -46,8 +48,9 @@
   "normalize_embedding": false,
   "num_beams": 6,
   "num_hidden_layers": 6,
-  "pad_token_id": 64683,
+  "pad_token_id": 63104,
   "scale_embedding": true,
   "static_position_embeddings": true,
-  "vocab_size": 64684
+  "use_cache": true,
+  "vocab_size": 63105
 }

metadata.json

@@ -0,0 +1 @@
+{"hf_name":"es-he","source_languages":"spa","target_languages":"heb","opus_readme_url":"https:\/\/github.com\/Helsinki-NLP\/Tatoeba-Challenge\/tree\/master\/models\/spa-heb\/README.md","original_repo":"Tatoeba-Challenge","tags":["translation"],"languages":["es","he"],"src_constituents":["Spanish",["spa"]],"tgt_constituents":["Hebrew",["heb"]],"src_multilingual":false,"tgt_multilingual":false,"long_pair":"spa-heb","prepro":" normalization + SentencePiece (spm32k,spm32k)","url_model":"https:\/\/object.pouta.csc.fi\/Tatoeba-MT-models\/spa-heb\/opus-2020-12-10.zip","url_test_set":"https:\/\/object.pouta.csc.fi\/Tatoeba-MT-models\/spa-heb\/opus-2020-12-10.test.txt","src_alpha3":"spa","tgt_alpha3":"heb","chrF2_score":0.636,"bleu":43.6,"brevity_penalty":0.992,"ref_len":12112.0,"src_name":"Spanish","tgt_name":"Hebrew","train_date":1607558400000,"src_alpha2":"es","tgt_alpha2":"he","prefer_old":false,"short_pair":"es-he","helsinki_git_sha":"b317f78a3ec8a556a481b6a53dc70dc11769ca96","transformers_git_sha":"1310e1a758edc8e89ec363db76863c771fbeb1de","port_machine":"LM0-400-22516.local","port_time":"2020-12-11-11:41"}

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:acb2be9ef637ae5abe44d66d8d1fb128ee382d9f816447cf3274885d55f54e3c
-size 311436525
+oid sha256:9b1d807ac7b402df12d461d58128b16e8c68ccd16c17a028663e501fed9b0dd4
+size 153110363

special_tokens_map.json

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

tokenizer_config.json

@@ -1 +1 @@
-{"target_lang": "he", "source_lang": "es"}
+{"source_lang": "spa", "target_lang": "heb", "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-heb"}