Upload 11 files

config.json

@@ -0,0 +1,24 @@
+{
+  "_name_or_path": "result/my-sup-simcse-bert-base-lastterm",
+  "architectures": [
+    "BertForCL"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "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,
+  "position_embedding_type": "absolute",
+  "transformers_version": "4.2.1",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

eval_results.txt

@@ -0,0 +1,12 @@
+epoch = 3.0
+eval_CR = 89.48
+eval_MPQA = 89.68
+eval_MR = 82.55
+eval_MRPC = 76.67
+eval_SST2 = 87.27
+eval_SUBJ = 94.48
+eval_TREC = 78.76
+eval_avg_sts = 0.8170722172992666
+eval_avg_transfer = 85.55571428571429
+eval_sickr_spearman = 0.7818400385849915
+eval_stsb_spearman = 0.8523043960135418

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df891a475e8137b73ce85bff69ebbc59b96f6bc0ef2ccf169ff9a9613e8605a8
+size 875973285

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16c5d3e54a49ce0c6c0a382525355ba9a6a5db0f9cb504bd1421ec54ed7a4eb3
+size 438012465

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17c4ae848e092d1dae85b8344f3b8f445b2c5373adfd9a7888568a422c676a4f
+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": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "name_or_path": "bert-base-uncased"}

train_results.txt

@@ -0,0 +1,3 @@
+epoch = 3.0
+train_runtime = 2076.5315
+train_samples_per_second = 1.556

trainer_state.json

@@ -0,0 +1,233 @@
+{
+  "best_metric": 0.8523043960135418,
+  "best_model_checkpoint": "result/my-sup-simcse-bert-base-lastterm",
+  "epoch": 3.0,
+  "global_step": 3231,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.12,
+      "eval_avg_sts": 0.8194254390928033,
+      "eval_sickr_spearman": 0.7990052495236399,
+      "eval_stsb_spearman": 0.8398456286619668,
+      "step": 125
+    },
+    {
+      "epoch": 0.23,
+      "eval_avg_sts": 0.8165670458735704,
+      "eval_sickr_spearman": 0.7869509320268182,
+      "eval_stsb_spearman": 0.8461831597203227,
+      "step": 250
+    },
+    {
+      "epoch": 0.35,
+      "eval_avg_sts": 0.8228159333151963,
+      "eval_sickr_spearman": 0.7957744855165066,
+      "eval_stsb_spearman": 0.8498573811138859,
+      "step": 375
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 4.226245744351594e-05,
+      "loss": 1.0808,
+      "step": 500
+    },
+    {
+      "epoch": 0.46,
+      "eval_avg_sts": 0.8168153761131349,
+      "eval_sickr_spearman": 0.7877662119426287,
+      "eval_stsb_spearman": 0.8458645402836412,
+      "step": 500
+    },
+    {
+      "epoch": 0.58,
+      "eval_avg_sts": 0.8156032349409434,
+      "eval_sickr_spearman": 0.7826891323962145,
+      "eval_stsb_spearman": 0.8485173374856724,
+      "step": 625
+    },
+    {
+      "epoch": 0.7,
+      "eval_avg_sts": 0.8169960567322327,
+      "eval_sickr_spearman": 0.7846676775577546,
+      "eval_stsb_spearman": 0.849324435906711,
+      "step": 750
+    },
+    {
+      "epoch": 0.81,
+      "eval_avg_sts": 0.8146560028725298,
+      "eval_sickr_spearman": 0.7835492733607682,
+      "eval_stsb_spearman": 0.8457627323842916,
+      "step": 875
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 3.452491488703188e-05,
+      "loss": 0.8522,
+      "step": 1000
+    },
+    {
+      "epoch": 0.93,
+      "eval_avg_sts": 0.8163927093002852,
+      "eval_sickr_spearman": 0.7824352399940127,
+      "eval_stsb_spearman": 0.8503501786065577,
+      "step": 1000
+    },
+    {
+      "epoch": 1.04,
+      "eval_avg_sts": 0.8052909347246509,
+      "eval_sickr_spearman": 0.7662991914658787,
+      "eval_stsb_spearman": 0.8442826779834229,
+      "step": 1125
+    },
+    {
+      "epoch": 1.16,
+      "eval_avg_sts": 0.8106314735517854,
+      "eval_sickr_spearman": 0.77574806186519,
+      "eval_stsb_spearman": 0.8455148852383809,
+      "step": 1250
+    },
+    {
+      "epoch": 1.28,
+      "eval_avg_sts": 0.8113568861098444,
+      "eval_sickr_spearman": 0.7783763237358834,
+      "eval_stsb_spearman": 0.8443374484838054,
+      "step": 1375
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 2.678737233054782e-05,
+      "loss": 0.6828,
+      "step": 1500
+    },
+    {
+      "epoch": 1.39,
+      "eval_avg_sts": 0.8208657443062924,
+      "eval_sickr_spearman": 0.7928466056350422,
+      "eval_stsb_spearman": 0.8488848829775427,
+      "step": 1500
+    },
+    {
+      "epoch": 1.51,
+      "eval_avg_sts": 0.8180871720303612,
+      "eval_sickr_spearman": 0.7858960729770345,
+      "eval_stsb_spearman": 0.8502782710836879,
+      "step": 1625
+    },
+    {
+      "epoch": 1.62,
+      "eval_avg_sts": 0.8170722172992666,
+      "eval_sickr_spearman": 0.7818400385849915,
+      "eval_stsb_spearman": 0.8523043960135418,
+      "step": 1750
+    },
+    {
+      "epoch": 1.74,
+      "eval_avg_sts": 0.81144903531131,
+      "eval_sickr_spearman": 0.7756615098203985,
+      "eval_stsb_spearman": 0.8472365608022215,
+      "step": 1875
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 1.9049829774063757e-05,
+      "loss": 0.6522,
+      "step": 2000
+    },
+    {
+      "epoch": 1.86,
+      "eval_avg_sts": 0.8141474203440927,
+      "eval_sickr_spearman": 0.7794347370871737,
+      "eval_stsb_spearman": 0.8488601036010117,
+      "step": 2000
+    },
+    {
+      "epoch": 1.97,
+      "eval_avg_sts": 0.8158956922741192,
+      "eval_sickr_spearman": 0.7808102517701584,
+      "eval_stsb_spearman": 0.8509811327780799,
+      "step": 2125
+    },
+    {
+      "epoch": 2.09,
+      "eval_avg_sts": 0.814587631326179,
+      "eval_sickr_spearman": 0.7796313764164613,
+      "eval_stsb_spearman": 0.8495438862358966,
+      "step": 2250
+    },
+    {
+      "epoch": 2.21,
+      "eval_avg_sts": 0.8114095673015015,
+      "eval_sickr_spearman": 0.7735022716552481,
+      "eval_stsb_spearman": 0.8493168629477549,
+      "step": 2375
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 1.1312287217579697e-05,
+      "loss": 0.5675,
+      "step": 2500
+    },
+    {
+      "epoch": 2.32,
+      "eval_avg_sts": 0.8121220784998155,
+      "eval_sickr_spearman": 0.7740748984465933,
+      "eval_stsb_spearman": 0.8501692585530379,
+      "step": 2500
+    },
+    {
+      "epoch": 2.44,
+      "eval_avg_sts": 0.8130334510314673,
+      "eval_sickr_spearman": 0.7768290018030546,
+      "eval_stsb_spearman": 0.84923790025988,
+      "step": 2625
+    },
+    {
+      "epoch": 2.55,
+      "eval_avg_sts": 0.8140933803428295,
+      "eval_sickr_spearman": 0.7770954303227319,
+      "eval_stsb_spearman": 0.8510913303629272,
+      "step": 2750
+    },
+    {
+      "epoch": 2.67,
+      "eval_avg_sts": 0.8149543004915998,
+      "eval_sickr_spearman": 0.7786629253181646,
+      "eval_stsb_spearman": 0.8512456756650348,
+      "step": 2875
+    },
+    {
+      "epoch": 2.79,
+      "learning_rate": 3.574744661095636e-06,
+      "loss": 0.5295,
+      "step": 3000
+    },
+    {
+      "epoch": 2.79,
+      "eval_avg_sts": 0.8162287995156676,
+      "eval_sickr_spearman": 0.7821816357784194,
+      "eval_stsb_spearman": 0.8502759632529158,
+      "step": 3000
+    },
+    {
+      "epoch": 2.9,
+      "eval_avg_sts": 0.8163857434466557,
+      "eval_sickr_spearman": 0.7826894686139245,
+      "eval_stsb_spearman": 0.850082018279387,
+      "step": 3125
+    },
+    {
+      "epoch": 3.0,
+      "step": 3231,
+      "train_runtime": 2076.5315,
+      "train_samples_per_second": 1.556
+    }
+  ],
+  "max_steps": 3231,
+  "num_train_epochs": 3,
+  "total_flos": 52139850005053440,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b54624ed6cd245ae35c94b90dbf8c13889510b6d52ad29eb3631dc50eb346b47
+size 2095