added model

0_Transformer/added_tokens.json

@@ -0,0 +1 @@
+{"IDcallcentreID": 119547, "IDsugardaddyID": 119548, "IDeconomicaidID": 119549, "IDsalacirúrgicaID": 119550, "IDdreamticketID": 119551, "IDancienthistoryID": 119552, "IDovelhanegraID": 119553, "IDweddinganniversaryID": 119554, "IDmailserviceID": 119555, "IDlabourunionID": 119556, "IDcellularphoneID": 119557, "IDsextosentidoID": 119558, "IDcomputerprogramID": 119559, "IDsanguequenteID": 119560, "IDmiddleschoolID": 119561, "IDconartistID": 119562, "IDsacredcowID": 119563, "IDsittingduckID": 119564, "IDbadhatID": 119565, "IDlifevestID": 119566, "IDpillowslipID": 119567, "IDbusybeeID": 119568, "IDfoodmarketID": 119569, "IDlivroabertoID": 119570, "IDcriticalreviewID": 119571, "IDprimeiradamaID": 119572, "IDvistagrossaID": 119573, "IDsinalverdeID": 119574, "IDloansharkID": 119575, "IDcentrocomercialID": 119576, "IDamigoocultoID": 119577, "IDpeaceconferenceID": 119578, "IDbananarepublicID": 119579, "IDmentaldisorderID": 119580, "IDmailinglistID": 119581, "IDweddingdayID": 119582, "IDprimeiranecessidadeID": 119583, "IDpublicserviceID": 119584, "IDbigfishID": 119585, "IDphonebookID": 119586, "IDfilaindianaID": 119587, "IDthinktankID": 119588, "IDswansongID": 119589, "IDgovernofederalID": 119590, "IDhighlifeID": 119591, "IDinsurancecompanyID": 119592, "IDbowtieID": 119593, "IDsegundoplanoID": 119594, "IDdiscovoadorID": 119595, "IDflowerchildID": 119596, "IDburaconegroID": 119597, "IDregimepolíticoID": 119598, "IDbraçodireitoID": 119599, "IDhomerunID": 119600, "IDanonovoID": 119601, "IDdisabilityinsuranceID": 119602, "IDbodeexpiatórioID": 119603, "IDcolégiomilitarID": 119604, "IDalarmefalsoID": 119605, "IDcottoncandyID": 119606, "IDchainreactionID": 119607, "IDtopdogID": 119608, "IDskintoneID": 119609, "IDroletarussaID": 119610, "IDnúcleoatômicoID": 119611, "IDfiocondutorID": 119612, "IDcaixeiroviajanteID": 119613, "IDsexbombID": 119614, "IDcomidacaseiraID": 119615, "IDdrylandID": 119616, "IDelbowroomID": 119617, "IDagentesecretoID": 119618, "IDtravelguideID": 119619, "IDricepaperID": 119620, "IDpontoforteID": 119621, "IDcookingstoveID": 119622, "IDbirthrateID": 119623, "IDpastoralemãoID": 119624, "IDbadappleID": 119625, "IDpandacarID": 119626, "IDhealthcheckID": 119627, "IDplantabaixaID": 119628, "IDdesfilemilitarID": 119629, "IDduststormID": 119630, "IDpapelhigiênicoID": 119631, "IDnarrowescapeID": 119632, "IDentrancehallID": 119633, "IDredesocialID": 119634, "IDcheiroverdeID": 119635, "IDgooseeggID": 119636, "IDprivateeyeID": 119637, "IDsanguefrioID": 119638, "IDaltatemporadaID": 119639, "IDfishstoryID": 119640, "IDsilverliningID": 119641, "IDclosedbookID": 119642, "IDleiteintegralID": 119643, "IDeagerbeaverID": 119644, "IDaltofalanteID": 119645, "IDprocessoseletivoID": 119646}

0_Transformer/config.json

@@ -0,0 +1,30 @@
+{
+  "_name_or_path": "models-no-git/mBERT-4",
+  "architectures": [
+    "BertModel"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "directionality": "bidi",
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "pooler_fc_size": 768,
+  "pooler_num_attention_heads": 12,
+  "pooler_num_fc_layers": 3,
+  "pooler_size_per_head": 128,
+  "pooler_type": "first_token_transform",
+  "position_embedding_type": "absolute",
+  "transformers_version": "4.7.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 119647
+}

0_Transformer/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:688f92a1aeb2dc73d7afcde0c1115b5daeb3f6b40015cea249fb3091b8722d00
+size 711805519

0_Transformer/sentence_bert_config.json

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

0_Transformer/special_tokens_map.json

@@ -0,0 +1 @@
+{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}

0_Transformer/tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": false, "do_basic_tokenize": true, "never_split": null, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "truncation": true, "model_max_length": 512, "special_tokens_map_file": null, "tokenizer_file": "/rds/homes/t/tayyarmh/.cache/huggingface/transformers/46880f3b0081fda494a4e15b05787692aa4c1e21e0ff2428ba8b14d4eda0784d.b33e51591f94f17c238ee9b1fac75b96ff2678cbaed6e108feadb3449d18dc24", "name_or_path": "models-no-git/mBERT-4", "max_length": 510}

1_Pooling/config.json

@@ -0,0 +1,7 @@
+{
+  "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
+}

config.json

@@ -0,0 +1,3 @@
+{
+  "__version__": "1.1.0"
+}

modules.json

@@ -0,0 +1,14 @@
+[
+  {
+    "idx": 0,
+    "name": "0",
+    "path": "0_Transformer",
+    "type": "sentence_transformers.models.Transformer"
+  },
+  {
+    "idx": 1,
+    "name": "1",
+    "path": "1_Pooling",
+    "type": "sentence_transformers.models.Pooling"
+  }
+]

similarity_evaluation_sts-dev_results.csv

@@ -0,0 +1,17 @@
+epoch,steps,cosine_pearson,cosine_spearman,euclidean_pearson,euclidean_spearman,manhattan_pearson,manhattan_spearman,dot_pearson,dot_spearman
+0,1000,0.8223683638890682,0.8193260746828027,0.7979897461315534,0.803220135294497,0.7972537041554002,0.8023219866647214,0.6836580185806979,0.6908144098854685
+0,2000,0.830664557640928,0.8328823326143419,0.8197270800357044,0.8261745039199623,0.8187195553775386,0.8253057420786962,0.7470980605382469,0.7539012192611184
+0,3000,0.8532102877932635,0.8538625293101365,0.8376568517743543,0.8472589430577713,0.8371536807203488,0.8466608863613057,0.7810536383543212,0.7810708149772704
+0,-1,0.8500315029829728,0.8578938984074234,0.8417480618030804,0.8511780956467101,0.8411655192026344,0.8506194087447622,0.7949301393810284,0.7924873890219402
+1,1000,0.8578620737935299,0.8582792092897007,0.8396138877006434,0.8494024974095088,0.8392435932491352,0.8490699491621757,0.8172337594749354,0.8109385741278216
+1,2000,0.8581231629987007,0.8593656560128075,0.8367399966979505,0.8484627454543238,0.8362490306241599,0.8480684523596608,0.8101933657847815,0.8064536983556769
+1,3000,0.8681338647595294,0.8696173391238481,0.8418244384908224,0.8530550465878121,0.8412889355836782,0.8527624753725964,0.8299749320088929,0.8263931846445596
+1,-1,0.8722190954710661,0.8724535222341263,0.8500635637337401,0.8599543104710159,0.8496808905785831,0.859633194782725,0.8275082318337017,0.8229208728033741
+2,1000,0.8713985764061338,0.873488982509562,0.8431598373071753,0.8561316748951959,0.84231541542985,0.8552921688276208,0.8266732100419826,0.8297623717778592
+2,2000,0.8752459707579081,0.8801139716896569,0.8498726030593595,0.8623087740718812,0.8488354359076971,0.8614224759686859,0.8382400695273857,0.8398862477265393
+2,3000,0.875848157073498,0.8807769458159065,0.8527219296659543,0.8654792909303771,0.8521607610527518,0.8652135910413792,0.8378492255919863,0.8374382824744216
+2,-1,0.8767945349920269,0.8808766350999973,0.8525646834342798,0.8654599356061631,0.851971596985079,0.8651104705244346,0.8382957913531548,0.8393058284962261
+3,1000,0.8791293176541888,0.8830838274821821,0.8500104117438607,0.8636591999362587,0.8495457877909478,0.8635084761509243,0.8459653650128598,0.8480190009427914
+3,2000,0.8805680423109419,0.8843310344914797,0.8515974005829676,0.8651675867045527,0.8510611851417531,0.8650140940987523,0.8442976277857414,0.8469192639753843
+3,3000,0.8818420961236912,0.8855876942484281,0.8515695700059771,0.8651456434236546,0.8510890986467661,0.8649132474482646,0.8463414101702343,0.8498252995950734
+3,-1,0.881865030205544,0.8856171269074611,0.8516142765740944,0.8652213248362883,0.8511326429138336,0.8649752674503688,0.8462780295211115,0.8497993907637191

similarity_evaluation_sts-test_results.csv

@@ -0,0 +1,2 @@
+epoch,steps,cosine_pearson,cosine_spearman,euclidean_pearson,euclidean_spearman,manhattan_pearson,manhattan_spearman,dot_pearson,dot_spearman
+-1,-1,0.8292630828165106,0.7998076959816071,0.8111914537624083,0.7957289259772087,0.811263624860446,0.7959313320750702,0.7583753141076951,0.6920700766848666