change base bert to kpfbert-sbert

README.md

@@ -85,7 +85,7 @@ The model was trained with the parameters:
 
 **DataLoader**:
 
-`torch.utils.data.dataloader.DataLoader` of length 3210 with parameters:
+`torch.utils.data.dataloader.DataLoader` of length 3001 with parameters:
 ```
 {'batch_size': 8, 'sampler': 'torch.utils.data.sampler.RandomSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'}
 ```
@@ -116,7 +116,7 @@ Parameters of the fit()-Method:
 ## Full Model Architecture
 ```
 SentenceTransformer(
-  (0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: BertModel 
+  (0): Transformer({'max_seq_length': 256, 'do_lower_case': True}) with Transformer model: BertModel 
   (1): Pooling({'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

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/root/.cache/torch/sentence_transformers/jhgan_ko-sbert-nli/",
+  "_name_or_path": "/root/.cache/torch/sentence_transformers/bongsoo_kpf-sbert-v1.1/",
   "architectures": [
     "BertModel"
   ],
@@ -17,9 +17,10 @@
   "num_hidden_layers": 12,
   "pad_token_id": 0,
   "position_embedding_type": "absolute",
+  "tokenizer_class": "BertTokenizerFast",
   "torch_dtype": "float32",
   "transformers_version": "4.26.1",
   "type_vocab_size": 2,
   "use_cache": true,
-  "vocab_size": 32000
+  "vocab_size": 36440
 }

config_sentence_transformers.json

@@ -1,7 +1,7 @@
 {
   "__version__": {
-    "sentence_transformers": "2.1.0",
-    "transformers": "4.13.0",
-    "pytorch": "1.7.0+cu110"
+    "sentence_transformers": "2.2.0",
+    "transformers": "4.21.2",
+    "pytorch": "1.10.1"
   }
 }

eval/similarity_evaluation_results.csv

@@ -1,7 +1,7 @@
 epoch,steps,cosine_pearson,cosine_spearman,euclidean_pearson,euclidean_spearman,manhattan_pearson,manhattan_spearman,dot_pearson,dot_spearman
--1,-1,0.9050268434700313,0.8061847764983627,0.8956832943172656,0.8053610602919355,0.8949142384752975,0.8051881524880684,0.8893832821171052,0.8037141627354802
-0,-1,0.9659270021124075,0.8142281374427333,0.9664289110515307,0.8141726391312473,0.9660866588814404,0.8140938461222601,0.9628513525963557,0.8136987110448793
-1,-1,0.9770033047332874,0.8154803066527413,0.9782559971035137,0.8155006651446879,0.9779874752518809,0.815483861312489,0.9752678152262114,0.8151033125756871
-2,-1,0.9797618845142864,0.8157571797789388,0.9830029174646199,0.8157736270371966,0.9827521505930031,0.8157725127139819,0.9787427532791765,0.8153935463036612
-3,-1,0.9818861491084675,0.8158954882024352,0.9858265803365402,0.8159058178876252,0.9855883584509971,0.8159033441003376,0.9810634940282239,0.8155765720632674
-4,-1,0.9819577188035005,0.8159573325261691,0.9866912274880311,0.815966737332761,0.9864590315489913,0.815962447218505,0.9813884007783665,0.8155776083753867
+-1,-1,0.7191056425052988,0.7076665465462761,0.5762445324673561,0.5813344414441548,0.573094893040687,0.578153904411696,0.6399484541152369,0.6432329065348992
+0,-1,0.9580800992357477,0.8138681389425984,0.9402369432937032,0.8143218508877417,0.9398479601224061,0.814281495430662,0.9080216800713337,0.8081089083981797
+1,-1,0.9736195473168833,0.815480807676057,0.9524687894002634,0.8154924143266667,0.9520028529401432,0.8154767134499858,0.9194615084828619,0.8110886566077198
+2,-1,0.9788963779981773,0.81565234389891,0.9608720685082192,0.8156816666421048,0.9604365533581654,0.8156825977266176,0.9310448601886931,0.8117031462928667
+3,-1,0.981274842167634,0.8157790988201438,0.9656924827365213,0.8158902293735341,0.9652933647637408,0.815891160455923,0.9339480889781077,0.811717150810645
+4,-1,0.9817855752729285,0.8158645289305935,0.9673852709598557,0.8159244243761873,0.967004821974988,0.8159262992965975,0.9343400044158627,0.8116573446311126

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:794936124001901ca20092d81c8b9edde41d2364586b15da068f7594cde31fcd
-size 442540589
+oid sha256:689e801ab8058d7586aaf2b8788af806ac767774cc63103f61a78d91a3238813
+size 456180269

sentence_bert_config.json

@@ -1,4 +1,4 @@
 {
-  "max_seq_length": 128,
-  "do_lower_case": false
+  "max_seq_length": 256,
+  "do_lower_case": true
 }

tokenizer_config.json

@@ -3,12 +3,12 @@
   "do_basic_tokenize": true,
   "do_lower_case": false,
   "mask_token": "[MASK]",
-  "model_max_length": 512,
-  "name_or_path": "/root/.cache/torch/sentence_transformers/jhgan_ko-sbert-nli/",
+  "model_max_length": 1000000000000000019884624838656,
+  "name_or_path": "/root/.cache/torch/sentence_transformers/bongsoo_kpf-sbert-v1.1/",
   "never_split": null,
   "pad_token": "[PAD]",
   "sep_token": "[SEP]",
-  "special_tokens_map_file": "/home/jhgan/.cache/huggingface/transformers/aeaaa3afd086a040be912f92ffe7b5f85008b744624f4517c4216bcc32b51cf0.054ece8d16bd524c8a00f0e8a976c00d5de22a755ffb79e353ee2954d9289e26",
+  "special_tokens_map_file": null,
   "strip_accents": null,
   "tokenize_chinese_chars": true,
   "tokenizer_class": "BertTokenizer",