Add onnx model export

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+onnx/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

1_Pooling/config.json

@@ -0,0 +1,10 @@
+{
+  "word_embedding_dimension": 768,
+  "pooling_mode_cls_token": false,
+  "pooling_mode_mean_tokens": true,
+  "pooling_mode_max_tokens": false,
+  "pooling_mode_mean_sqrt_len_tokens": false,
+  "pooling_mode_weightedmean_tokens": false,
+  "pooling_mode_lasttoken": false,
+  "include_prompt": true
+}

config.json

@@ -0,0 +1,27 @@
+{
+  "architectures": [
+    "XLMRobertaModel"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "xlm-roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.50.3",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 250002
+}

config_sentence_transformers.json

@@ -0,0 +1,13 @@
+{
+  "__version__": {
+    "sentence_transformers": "4.0.1",
+    "transformers": "4.50.3",
+    "pytorch": "2.6.0+cu124"
+  },
+  "prompts": {
+    "query": "query: ",
+    "text": "passage: "
+  },
+  "default_prompt_name": null,
+  "similarity_fn_name": "dot"
+}

eval/Information-Retrieval_evaluation_jargonisms-ru--matryoshka_dim-768--_results.csv

@@ -0,0 +1,4 @@
+epoch,steps,dot-Accuracy@10,dot-Precision@10,dot-Recall@10,dot-MRR@10,dot-NDCG@10,dot-MAP@60
+0.3333333333333333,1,0.9743589743589743,0.6179487179487179,0.4105273910765902,0.9081196581196581,0.7489199118227761,0.4309364769470105
+0.6666666666666666,2,0.9743589743589743,0.6153846153846154,0.40796328851248764,0.8867521367521368,0.7411956524794387,0.4262143644881094
+1.0,3,0.9743589743589743,0.6153846153846154,0.40796328851248764,0.8952991452991454,0.7456236376148996,0.43053013067186624

eval/Information-Retrieval_evaluation_jargonisms-uk--matryoshka_dim-768--_results.csv

@@ -0,0 +1,4 @@
+epoch,steps,dot-Accuracy@10,dot-Precision@10,dot-Recall@10,dot-MRR@10,dot-NDCG@10,dot-MAP@60
+0.3333333333333333,1,0.9743589743589743,0.6282051282051282,0.4175795981287973,0.901098901098901,0.7588065127999831,0.4400081713196617
+0.6666666666666666,2,0.9743589743589743,0.6256410256410257,0.4150154955646947,0.9096459096459097,0.7601604035409352,0.44351847499359603
+1.0,3,0.9743589743589743,0.6282051282051282,0.4175795981287973,0.901098901098901,0.7586242128347274,0.44004619209370444

eval/Information-Retrieval_evaluation_options-ru--matryoshka_dim-768--_results.csv

@@ -0,0 +1,4 @@
+epoch,steps,dot-Accuracy@10,dot-Precision@10,dot-Recall@10,dot-MRR@10,dot-NDCG@10,dot-MAP@60
+0.3333333333333333,1,0.9927184466019418,0.34126213592233,0.48700444983818775,0.8831445523193094,0.5759040356170075,0.415424507059974
+0.6666666666666666,2,0.9927184466019418,0.3405339805825242,0.48625606796116505,0.8893974418246262,0.5769810386224125,0.41666669958389974
+1.0,3,0.9927184466019418,0.3405339805825243,0.4860942556634304,0.8825204191709047,0.5754815076749632,0.4155437503228611

eval/Information-Retrieval_evaluation_options-uk--matryoshka_dim-768--_results.csv

@@ -0,0 +1,4 @@
+epoch,steps,dot-Accuracy@10,dot-Precision@10,dot-Recall@10,dot-MRR@10,dot-NDCG@10,dot-MAP@60
+0.3333333333333333,1,0.9951456310679612,0.3390776699029126,0.4844356796116504,0.8896199337340114,0.5761432869978054,0.4166583720837996
+0.6666666666666666,2,0.9951456310679612,0.3393203883495145,0.48484021035598707,0.8905272384034519,0.5760489662024446,0.41610360224293846
+1.0,3,0.9951456310679612,0.3393203883495146,0.4850424757281553,0.8889495685005392,0.5754311792342823,0.4155063286471065

eval/Information-Retrieval_evaluation_rusisms-ru--matryoshka_dim-768--_results.csv

@@ -0,0 +1,4 @@
+epoch,steps,dot-Accuracy@10,dot-Precision@10,dot-Recall@10,dot-MRR@10,dot-NDCG@10,dot-MAP@60
+0.3333333333333333,1,1.0,0.7263736263736265,0.3791388157153021,0.9527472527472528,0.8282440602910573,0.4750503350134536
+0.6666666666666666,2,1.0,0.7263736263736265,0.3791562585898878,0.9521193092621664,0.8280053809981986,0.47497714151829123
+1.0,3,1.0,0.7274725274725276,0.3794179017086738,0.9542124542124542,0.8291269750230967,0.47537466501489184

eval/Information-Retrieval_evaluation_rusisms-uk--matryoshka_dim-768--_results.csv

@@ -0,0 +1,4 @@
+epoch,steps,dot-Accuracy@10,dot-Precision@10,dot-Recall@10,dot-MRR@10,dot-NDCG@10,dot-MAP@60
+0.3333333333333333,1,1.0,0.724175824175824,0.3784346412716318,0.9645909645909645,0.8278567516600803,0.47536177172516614
+0.6666666666666666,2,1.0,0.7241758241758243,0.3784634989685568,0.9642857142857143,0.8288102091982976,0.4760882899463229
+1.0,3,1.0,0.7252747252747253,0.3787780478650384,0.9652014652014653,0.8292463278748027,0.4759284584016334

eval/Information-Retrieval_evaluation_test-core-ru--matryoshka_dim-768--_results.csv

@@ -0,0 +1,4 @@
+epoch,steps,dot-Accuracy@10,dot-Precision@10,dot-Recall@10,dot-MRR@10,dot-NDCG@10,dot-MAP@60
+0.3333333333333333,1,0.9934383202099738,0.40209973753280837,0.4786883410407033,0.8958197933591635,0.5997346112835937,0.43202948404936903
+0.6666666666666666,2,0.9921259842519685,0.402755905511811,0.47935544515268935,0.8988470191226098,0.6011798359830502,0.4330015110893986
+1.0,3,0.9934383202099738,0.40275590551181106,0.4794346019247594,0.8988668083156274,0.6011492872857404,0.43298615186536105

eval/Information-Retrieval_evaluation_test-core-uk--matryoshka_dim-768--_results.csv

@@ -0,0 +1,7 @@
+epoch,steps,dot-Accuracy@10,dot-Precision@10,dot-Recall@10,dot-MRR@10,dot-NDCG@10,dot-MAP@60
+0.5,1,0.9803149606299213,0.3796587926509187,0.45338942007249095,0.8448881389826275,0.5556007249636264,0.39181454287940987
+1.0,2,0.9763779527559056,0.38005249343832026,0.4532149627129942,0.8474721909761281,0.5582504927869466,0.39547573316621226
+0.3333333333333333,1,0.9934383202099738,0.4026246719160104,0.47908100029163025,0.8980257676123816,0.599329538563686,0.43047437097811636
+0.3333333333333333,1,0.9934383202099738,0.4026246719160104,0.47908100029163025,0.8980257676123816,0.599329538563686,0.43047437097811636
+0.6666666666666666,2,0.9934383202099738,0.4034120734908137,0.48011368370620333,0.8953859934174896,0.5999755578107874,0.43115499070142066
+1.0,3,0.9921259842519685,0.402755905511811,0.4790804795233929,0.8987621338999292,0.599688674233464,0.4309445991968187

eval/Information-Retrieval_evaluation_transliterations-ru--matryoshka_dim-768--_results.csv

@@ -0,0 +1,4 @@
+epoch,steps,dot-Accuracy@10,dot-Precision@10,dot-Recall@10,dot-MRR@10,dot-NDCG@10,dot-MAP@60
+0.3333333333333333,1,0.8508771929824561,0.20126230209670518,0.39328478717118,0.7644660431566723,0.47640653918405973,0.3649715069265092
+0.6666666666666666,2,0.8468121523320497,0.19895164741121096,0.39010139678464456,0.758758260828223,0.4726916566896715,0.362378577344258
+1.0,3,0.8480958493795464,0.19910141206675228,0.3907096246714347,0.7601083501436527,0.4733132151063623,0.3628110633317572

eval/Information-Retrieval_evaluation_transliterations-uk--matryoshka_dim-768--_results.csv

@@ -0,0 +1,4 @@
+epoch,steps,dot-Accuracy@10,dot-Precision@10,dot-Recall@10,dot-MRR@10,dot-NDCG@10,dot-MAP@60
+0.3333333333333333,1,0.8472400513478819,0.1999572100984168,0.39065359027650426,0.7584839469948588,0.4727749412177926,0.3616232194645004
+0.6666666666666666,2,0.8440308087291399,0.19758237056054773,0.387620601299998,0.752198161392642,0.4683243042486316,0.3580769931996699
+1.0,3,0.8429610611895593,0.19794608472400516,0.38779367116979446,0.7534757455970555,0.4689830934527895,0.3586187856349942

eval/Information-Retrieval_evaluation_typos-ru--matryoshka_dim-768--_results.csv

@@ -0,0 +1,4 @@
+epoch,steps,dot-Accuracy@10,dot-Precision@10,dot-Recall@10,dot-MRR@10,dot-NDCG@10,dot-MAP@60
+0.3333333333333333,1,0.9763779527559056,0.3803149606299212,0.4530009269674624,0.8352476773736613,0.5573643772950146,0.39792598467117285
+0.6666666666666666,2,0.9724409448818898,0.38123359580052496,0.45333343748698074,0.8383118776819564,0.558472232653569,0.39854206356359606
+1.0,3,0.9763779527559056,0.3809711286089239,0.45344748573095023,0.8357252218472689,0.5576777101925224,0.3979690284194279

eval/Information-Retrieval_evaluation_typos-uk--matryoshka_dim-768--_results.csv

@@ -0,0 +1,4 @@
+epoch,steps,dot-Accuracy@10,dot-Precision@10,dot-Recall@10,dot-MRR@10,dot-NDCG@10,dot-MAP@60
+0.3333333333333333,1,0.973753280839895,0.381496062992126,0.45312669249677123,0.8343446652501773,0.5572985692961213,0.3969871056601361
+0.6666666666666666,2,0.973753280839895,0.383989501312336,0.45586957880264967,0.8408797858601008,0.5611968779000402,0.39931545679895825
+1.0,3,0.9724409448818898,0.38346456692913394,0.45519830854476523,0.8384311336082991,0.5603339981099429,0.39889596717888554

eval/Information-Retrieval_evaluation_vespa-ru--matryoshka_dim-768--_results.csv

@@ -0,0 +1,4 @@
+epoch,steps,dot-Accuracy@10,dot-Precision@10,dot-Recall@10,dot-MRR@10,dot-NDCG@10,dot-MAP@60
+0.3333333333333333,1,0.9239130434782609,0.6891304347826087,0.2347958205099316,0.8558100414078674,0.7650142033568607,0.5636621361346738
+0.6666666666666666,2,0.9239130434782609,0.6869565217391305,0.23447565607291931,0.8567934782608695,0.762999185508784,0.563098096767067
+1.0,3,0.9239130434782609,0.691304347826087,0.23597022129031064,0.8540760869565216,0.7659189610978123,0.5637191455131668

eval/Information-Retrieval_evaluation_vespa-uk--matryoshka_dim-768--_results.csv

@@ -0,0 +1,4 @@
+epoch,steps,dot-Accuracy@10,dot-Precision@10,dot-Recall@10,dot-MRR@10,dot-NDCG@10,dot-MAP@60
+0.3333333333333333,1,0.9130434782608695,0.6869565217391305,0.23873817086144347,0.8702294685990337,0.7712268265158239,0.5737831004248664
+0.6666666666666666,2,0.9021739130434783,0.6869565217391305,0.23512468742459045,0.8644927536231883,0.7678413987836414,0.5726721927660509
+1.0,3,0.9130434782608695,0.6858695652173914,0.2394267615150962,0.8641304347826086,0.7686399855634083,0.5724795153965255

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:563bb107577c51458456b6b0981eb8218c5744d91e63933121be3dfb1ebc22ac
+size 556109872

modules.json

@@ -0,0 +1,20 @@
+[
+  {
+    "idx": 0,
+    "name": "0",
+    "path": "",
+    "type": "sentence_transformers.models.Transformer"
+  },
+  {
+    "idx": 1,
+    "name": "1",
+    "path": "1_Pooling",
+    "type": "sentence_transformers.models.Pooling"
+  },
+  {
+    "idx": 2,
+    "name": "2",
+    "path": "2_Normalize",
+    "type": "sentence_transformers.models.Normalize"
+  }
+]

onnx/config.json

@@ -0,0 +1,29 @@
+{
+  "_attn_implementation_autoset": true,
+  "architectures": [
+    "XLMRobertaModel"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "export_model_type": "transformer",
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "xlm-roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.50.3",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 250002
+}

onnx/model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81f7edbc38aca825d6e3ef91864146c3ac6920a300cd72b625e86d2369159b6e
+size 1110083307

onnx/sentencepiece.bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
+size 5069051

onnx/special_tokens_map.json

@@ -0,0 +1,51 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "cls_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

onnx/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:883b037111086fd4dfebbbc9b7cee11e1517b5e0c0514879478661440f137085
+size 17082987

onnx/tokenizer_config.json

@@ -0,0 +1,62 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250001": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "extra_special_tokens": {},
+  "mask_token": "<mask>",
+  "max_length": 512,
+  "model_max_length": 512,
+  "pad_to_multiple_of": null,
+  "pad_token": "<pad>",
+  "pad_token_type_id": 0,
+  "padding_side": "right",
+  "sep_token": "</s>",
+  "stride": 0,
+  "tokenizer_class": "XLMRobertaTokenizer",
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first",
+  "unk_token": "<unk>"
+}

sentence_bert_config.json

@@ -0,0 +1,4 @@
+{
+  "max_seq_length": 512,
+  "do_lower_case": false
+}

sentencepiece.bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
+size 5069051

special_tokens_map.json

@@ -0,0 +1,51 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "cls_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:883b037111086fd4dfebbbc9b7cee11e1517b5e0c0514879478661440f137085
+size 17082987

tokenizer_config.json

@@ -0,0 +1,62 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250001": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "extra_special_tokens": {},
+  "mask_token": "<mask>",
+  "max_length": 512,
+  "model_max_length": 512,
+  "pad_to_multiple_of": null,
+  "pad_token": "<pad>",
+  "pad_token_type_id": 0,
+  "padding_side": "right",
+  "sep_token": "</s>",
+  "stride": 0,
+  "tokenizer_class": "XLMRobertaTokenizerFast",
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first",
+  "unk_token": "<unk>"
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc638c6504d821de88fabdd84c289c9b10418c4a65f2a5d7b2b979ca77d6daa6
+size 5944