test

config.json

@@ -0,0 +1,30 @@
+{
+  "_name_or_path": "result/sup-simcse-nb-bert-base-gpu",
+  "architectures": [
+    "BertForCL"
+  ],
+  "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": "4.2.1",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 119547
+}

eval_results.txt

@@ -0,0 +1,12 @@
+epoch = 3.0
+eval_CR = 87.24
+eval_MPQA = 86.71
+eval_MR = 77.53
+eval_MRPC = 75.27
+eval_SST2 = 82.57
+eval_SUBJ = 92.92
+eval_TREC = 80.65
+eval_avg_sts = 0.742895730622507
+eval_avg_transfer = 83.27
+eval_sickr_spearman = 0.6944995319003601
+eval_stsb_spearman = 0.7912919293446539

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d0d0260de232f8fb1916af622417338f3bf3ba3ecb004cba66c043acba80e646
+size 1422952429

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c3c10e1d2f714c81e7f5227c8c620312061550bbf5b105f65928f85f057e4047
+size 711502455

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a50fec66ca2db0ebb243dfe7735bd83c0ac8399e49abd748b31c3a6392c01296
+size 623

special_tokens_map.json

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

tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": false, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "special_tokens_map_file": null, "name_or_path": "NbAiLab/nb-bert-base", "do_basic_tokenize": true, "never_split": null}

train_results.txt

@@ -0,0 +1,3 @@
+epoch = 3.0
+train_runtime = 1109.7525
+train_samples_per_second = 1.354

trainer_state.json

@@ -0,0 +1,124 @@
+{
+  "best_metric": 0.7912919293446539,
+  "best_model_checkpoint": "result/sup-simcse-nb-bert-base-gpu",
+  "epoch": 3.0,
+  "global_step": 1503,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.25,
+      "eval_avg_sts": 0.7442844791361708,
+      "eval_sickr_spearman": 0.6979807614334373,
+      "eval_stsb_spearman": 0.7905881968389042,
+      "step": 125
+    },
+    {
+      "epoch": 0.5,
+      "eval_avg_sts": 0.7342303358293948,
+      "eval_sickr_spearman": 0.6795624670893701,
+      "eval_stsb_spearman": 0.7888982045694195,
+      "step": 250
+    },
+    {
+      "epoch": 0.75,
+      "eval_avg_sts": 0.7320311452495603,
+      "eval_sickr_spearman": 0.6800595572796481,
+      "eval_stsb_spearman": 0.7840027332194723,
+      "step": 375
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.33666001330672e-05,
+      "loss": 0.61,
+      "step": 500
+    },
+    {
+      "epoch": 1.0,
+      "eval_avg_sts": 0.7362036208318414,
+      "eval_sickr_spearman": 0.6922774043553716,
+      "eval_stsb_spearman": 0.7801298373083113,
+      "step": 500
+    },
+    {
+      "epoch": 1.25,
+      "eval_avg_sts": 0.7370200639699311,
+      "eval_sickr_spearman": 0.6899311330812212,
+      "eval_stsb_spearman": 0.7841089948586409,
+      "step": 625
+    },
+    {
+      "epoch": 1.5,
+      "eval_avg_sts": 0.742895730622507,
+      "eval_sickr_spearman": 0.6944995319003601,
+      "eval_stsb_spearman": 0.7912919293446539,
+      "step": 750
+    },
+    {
+      "epoch": 1.75,
+      "eval_avg_sts": 0.7385084843412628,
+      "eval_sickr_spearman": 0.6870715534260018,
+      "eval_stsb_spearman": 0.7899454152565237,
+      "step": 875
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.67332002661344e-05,
+      "loss": 0.3372,
+      "step": 1000
+    },
+    {
+      "epoch": 2.0,
+      "eval_avg_sts": 0.7360397443568905,
+      "eval_sickr_spearman": 0.6866957100572486,
+      "eval_stsb_spearman": 0.7853837786565324,
+      "step": 1000
+    },
+    {
+      "epoch": 2.25,
+      "eval_avg_sts": 0.7377985562490148,
+      "eval_sickr_spearman": 0.6860993078706915,
+      "eval_stsb_spearman": 0.7894978046273381,
+      "step": 1125
+    },
+    {
+      "epoch": 2.5,
+      "eval_avg_sts": 0.7330533694034538,
+      "eval_sickr_spearman": 0.6807181434210725,
+      "eval_stsb_spearman": 0.785388595385835,
+      "step": 1250
+    },
+    {
+      "epoch": 2.74,
+      "eval_avg_sts": 0.7370673138097181,
+      "eval_sickr_spearman": 0.6858417651047803,
+      "eval_stsb_spearman": 0.7882928625146558,
+      "step": 1375
+    },
+    {
+      "epoch": 2.99,
+      "learning_rate": 9.98003992015968e-08,
+      "loss": 0.2239,
+      "step": 1500
+    },
+    {
+      "epoch": 2.99,
+      "eval_avg_sts": 0.7375278891429025,
+      "eval_sickr_spearman": 0.6869925422641361,
+      "eval_stsb_spearman": 0.7880632360216687,
+      "step": 1500
+    },
+    {
+      "epoch": 3.0,
+      "step": 1503,
+      "train_runtime": 1109.7525,
+      "train_samples_per_second": 1.354
+    }
+  ],
+  "max_steps": 1503,
+  "num_train_epochs": 3,
+  "total_flos": 39364151055482880,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f38dbce501e29db55bd2a3bfaf55a2c1dbde1d9a8b16d942c55984c3ded269a5
+size 2031