add weights

CEF1Evaluator_NLI-validation_results.csv

@@ -0,0 +1,46 @@
+epoch,steps,Macro F1 score,Micro F1 score,Weighted F1 score
+0,2000,0.8329982997836369,0.8375893769152196,0.8397220994618456
+0,4000,0.8748631523421083,0.877170582226762,0.8781035228038937
+0,6000,0.8820278411731026,0.8845760980592441,0.8863110758809742
+0,8000,0.9044091342906085,0.9057711950970377,0.9062963985693472
+0,10000,0.9107679086815503,0.9116445352400409,0.9119410291654654
+0,12000,0.9115665127927763,0.9126659856996936,0.9129618799913063
+0,14000,0.918864685145166,0.9195607763023493,0.919858180337296
+0,16000,0.9216040314629798,0.9218590398365679,0.9219611176600517
+0,-1,0.9100076248301333,0.9113891726251276,0.9120817353487438
+1,2000,0.9194390956139591,0.9200715015321757,0.9205939217972634
+1,4000,0.9185142962179245,0.9195607763023493,0.920105679030943
+1,6000,0.9208509751453576,0.9218590398365679,0.9223203005095948
+1,8000,0.9237640789148914,0.9244126659856997,0.9247183179540795
+1,10000,0.9280237475657899,0.9287538304392237,0.9290009296396736
+1,12000,0.9290262560148626,0.9300306435137896,0.9303821739012206
+1,14000,0.9278119404085364,0.9284984678243106,0.9285618797839071
+1,16000,0.928274634560662,0.9287538304392237,0.9288574304035064
+1,-1,0.9292224965106985,0.9297752808988764,0.9298854364884377
+2,2000,0.9317667502094175,0.9323289070480082,0.9326311979688516
+2,4000,0.9316300064298523,0.932073544433095,0.9321788837608004
+2,6000,0.9339594129221932,0.9343718079673136,0.9345856616322523
+2,8000,0.9344685355631314,0.93488253319714,0.9351301368674186
+2,10000,0.9328423146776924,0.933605720122574,0.9340127392617862
+2,12000,0.9285830287395637,0.9290091930541369,0.9292837395999622
+2,14000,0.931181352354283,0.9315628192032687,0.9317430287548752
+2,16000,0.9353571804783342,0.9359039836567926,0.9361475549571838
+2,-1,0.9323081476150016,0.9330949948927477,0.9334404916701219
+3,2000,0.9348641756711965,0.9356486210418795,0.9360014132732422
+3,4000,0.9300953744501271,0.9307967313585291,0.9310883640460593
+3,6000,0.9281845389858999,0.9284984678243106,0.9286734388514554
+3,8000,0.9324754511309039,0.9330949948927477,0.9333753646312469
+3,10000,0.9329047150272097,0.9333503575076609,0.9336977325814476
+3,12000,0.9333746327956094,0.9338610827374872,0.9341597867262501
+3,14000,0.9343758829846527,0.93488253319714,0.9351112846933961
+3,16000,0.9319437286185271,0.9325842696629213,0.9328579411370967
+3,-1,0.9325021374847483,0.9330949948927477,0.9334258830555647
+4,2000,0.931510475622838,0.9318181818181818,0.9319933144454969
+4,4000,0.933075533127825,0.933605720122574,0.9339175403522527
+4,6000,0.9318318095001942,0.9323289070480082,0.9326460598298444
+4,8000,0.930807201375508,0.9313074565883555,0.9316409459113678
+4,10000,0.9288952367942497,0.9292645556690501,0.9295356054499537
+4,12000,0.9301385094861797,0.9305413687436159,0.9308744533589931
+4,14000,0.9314496565220546,0.9318181818181818,0.9321588646101233
+4,16000,0.9303977405404575,0.9307967313585291,0.931146034919378
+4,-1,0.9309454479593707,0.9313074565883555,0.9316510303699307

CESoftmaxAccuracyEvaluator_NLI-validation_results.csv

@@ -0,0 +1,46 @@
+epoch,steps,Accuracy
+0,2000,0.8375893769152196
+0,4000,0.877170582226762
+0,6000,0.8845760980592441
+0,8000,0.9057711950970377
+0,10000,0.9116445352400409
+0,12000,0.9126659856996936
+0,14000,0.9195607763023493
+0,16000,0.9218590398365679
+0,-1,0.9113891726251276
+1,2000,0.9200715015321757
+1,4000,0.9195607763023493
+1,6000,0.9218590398365679
+1,8000,0.9244126659856997
+1,10000,0.9287538304392237
+1,12000,0.9300306435137896
+1,14000,0.9284984678243106
+1,16000,0.9287538304392237
+1,-1,0.9297752808988764
+2,2000,0.9323289070480082
+2,4000,0.932073544433095
+2,6000,0.9343718079673136
+2,8000,0.93488253319714
+2,10000,0.933605720122574
+2,12000,0.9290091930541369
+2,14000,0.9315628192032687
+2,16000,0.9359039836567926
+2,-1,0.9330949948927477
+3,2000,0.9356486210418795
+3,4000,0.9307967313585291
+3,6000,0.9284984678243106
+3,8000,0.9330949948927477
+3,10000,0.9333503575076609
+3,12000,0.9338610827374872
+3,14000,0.93488253319714
+3,16000,0.9325842696629213
+3,-1,0.9330949948927477
+4,2000,0.9318181818181818
+4,4000,0.933605720122574
+4,6000,0.9323289070480082
+4,8000,0.9313074565883555
+4,10000,0.9292645556690501
+4,12000,0.9305413687436159
+4,14000,0.9318181818181818
+4,16000,0.9307967313585291
+4,-1,0.9313074565883555

config.json

@@ -0,0 +1,35 @@
+{
+  "_name_or_path": "tohoku-nlp/bert-base-japanese-v3",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "entailment",
+    "1": "natural",
+    "2": "contradiction"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "entailment": 0,
+    "natural": 1,
+    "contradiction": 2
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.39.2",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 32768
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8189659e235adcd7bf0f079dab4443bb9afc50cf2c7592727d3d98977230d11c
+size 444861444

settings.json

@@ -0,0 +1,14 @@
+{
+    "output_dir": "./output/tohoku_bert_v3_v0.1",
+    "datasets": [
+        "shunk031/jsnli"
+    ],
+    "model": "tohoku-nlp/bert-base-japanese-v3",
+    "eval_steps": 2000,
+    "batch_size": 32,
+    "num_epochs": 5,
+    "warmup_ratio": 0.1,
+    "weight_decay": 0.01,
+    "show_progress_bar": true,
+    "scheduler": "warmupcosine"
+}

special_tokens_map.json

@@ -0,0 +1,7 @@
+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

tokenizer_config.json

@@ -0,0 +1,63 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_lower_case": false,
+  "do_subword_tokenize": true,
+  "do_word_tokenize": true,
+  "jumanpp_kwargs": null,
+  "mask_token": "[MASK]",
+  "mecab_kwargs": {
+    "mecab_dic": "unidic_lite"
+  },
+  "model_max_length": 512,
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "subword_tokenizer_type": "wordpiece",
+  "sudachi_kwargs": null,
+  "tokenizer_class": "BertJapaneseTokenizer",
+  "unk_token": "[UNK]",
+  "word_tokenizer_type": "mecab"
+}