End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 tags:
 - generated_from_trainer
 datasets:
@@ -12,7 +14,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE STSB
       type: glue
       config: stsb
       split: validation
@@ -20,7 +22,7 @@ model-index:
     metrics:
     - name: Spearmanr
       type: spearmanr
-      value: 0.3868434767975089
+      value: 0.4028161409951644
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -28,12 +30,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hBERTv2_new_pretrain_48_stsb
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new_48](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new_48) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new_48](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new_48) on the GLUE STSB dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.3369
-- Pearson: 0.4033
-- Spearmanr: 0.3868
-- Combined Score: 0.3951
+- Loss: 2.0734
+- Pearson: 0.4184
+- Spearmanr: 0.4028
+- Combined Score: 0.4106
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 19.0,
+    "eval_combined_score": 0.4106013077457039,
+    "eval_loss": 2.0734338760375977,
+    "eval_pearson": 0.41838647449624344,
+    "eval_runtime": 2.4898,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 602.446,
+    "eval_spearmanr": 0.4028161409951644,
+    "eval_steps_per_second": 4.82,
+    "train_loss": 0.7103834141067594,
+    "train_runtime": 695.8903,
+    "train_samples": 5749,
+    "train_samples_per_second": 413.068,
+    "train_steps_per_second": 3.233
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 19.0,
+    "eval_combined_score": 0.4106013077457039,
+    "eval_loss": 2.0734338760375977,
+    "eval_pearson": 0.41838647449624344,
+    "eval_runtime": 2.4898,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 602.446,
+    "eval_spearmanr": 0.4028161409951644,
+    "eval_steps_per_second": 4.82
+}

logs/events.out.tfevents.1686054958.serv-9207.2387813.20

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:343e7eadab812a93698ed9da0af2063b852af394738b084d0fd5c8b535e55a22
+size 473

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 19.0,
+    "train_loss": 0.7103834141067594,
+    "train_runtime": 695.8903,
+    "train_samples": 5749,
+    "train_samples_per_second": 413.068,
+    "train_steps_per_second": 3.233
+}

trainer_state.json

@@ -0,0 +1,348 @@
+{
+  "best_metric": 2.0734338760375977,
+  "best_model_checkpoint": "hBERTv2_new_pretrain_48_stsb/checkpoint-630",
+  "epoch": 19.0,
+  "global_step": 855,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9200000000000004e-05,
+      "loss": 2.2864,
+      "step": 45
+    },
+    {
+      "epoch": 1.0,
+      "eval_combined_score": 0.12204607317221586,
+      "eval_loss": 3.015711545944214,
+      "eval_pearson": 0.12695010020402245,
+      "eval_runtime": 2.4906,
+      "eval_samples_per_second": 602.273,
+      "eval_spearmanr": 0.11714204614040925,
+      "eval_steps_per_second": 4.818,
+      "step": 45
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 1.9895,
+      "step": 90
+    },
+    {
+      "epoch": 2.0,
+      "eval_combined_score": 0.15516534362604684,
+      "eval_loss": 2.7269980907440186,
+      "eval_pearson": 0.15532406574343285,
+      "eval_runtime": 2.4713,
+      "eval_samples_per_second": 606.959,
+      "eval_spearmanr": 0.15500662150866085,
+      "eval_steps_per_second": 4.856,
+      "step": 90
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.76e-05,
+      "loss": 1.7101,
+      "step": 135
+    },
+    {
+      "epoch": 3.0,
+      "eval_combined_score": 0.2731764320376846,
+      "eval_loss": 2.82230281829834,
+      "eval_pearson": 0.2806107701126286,
+      "eval_runtime": 2.4777,
+      "eval_samples_per_second": 605.408,
+      "eval_spearmanr": 0.2657420939627406,
+      "eval_steps_per_second": 4.843,
+      "step": 135
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 1.2973,
+      "step": 180
+    },
+    {
+      "epoch": 4.0,
+      "eval_combined_score": 0.3327597028880893,
+      "eval_loss": 2.5938479900360107,
+      "eval_pearson": 0.3375229154560037,
+      "eval_runtime": 2.4919,
+      "eval_samples_per_second": 601.946,
+      "eval_spearmanr": 0.32799649032017486,
+      "eval_steps_per_second": 4.816,
+      "step": 180
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.6e-05,
+      "loss": 1.0658,
+      "step": 225
+    },
+    {
+      "epoch": 5.0,
+      "eval_combined_score": 0.3700023404239433,
+      "eval_loss": 2.3834636211395264,
+      "eval_pearson": 0.377092889033301,
+      "eval_runtime": 2.5195,
+      "eval_samples_per_second": 595.344,
+      "eval_spearmanr": 0.36291179181458566,
+      "eval_steps_per_second": 4.763,
+      "step": 225
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 3.52e-05,
+      "loss": 0.8454,
+      "step": 270
+    },
+    {
+      "epoch": 6.0,
+      "eval_combined_score": 0.355781439962202,
+      "eval_loss": 2.5028064250946045,
+      "eval_pearson": 0.363688137161137,
+      "eval_runtime": 2.4788,
+      "eval_samples_per_second": 605.121,
+      "eval_spearmanr": 0.34787474276326696,
+      "eval_steps_per_second": 4.841,
+      "step": 270
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 3.44e-05,
+      "loss": 0.6773,
+      "step": 315
+    },
+    {
+      "epoch": 7.0,
+      "eval_combined_score": 0.35661381040781137,
+      "eval_loss": 2.393719434738159,
+      "eval_pearson": 0.3594036490491602,
+      "eval_runtime": 2.4803,
+      "eval_samples_per_second": 604.763,
+      "eval_spearmanr": 0.3538239717664625,
+      "eval_steps_per_second": 4.838,
+      "step": 315
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 0.5678,
+      "step": 360
+    },
+    {
+      "epoch": 8.0,
+      "eval_combined_score": 0.3802876739605978,
+      "eval_loss": 2.6813249588012695,
+      "eval_pearson": 0.38033678117885245,
+      "eval_runtime": 2.4841,
+      "eval_samples_per_second": 603.845,
+      "eval_spearmanr": 0.38023856674234324,
+      "eval_steps_per_second": 4.831,
+      "step": 360
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 3.28e-05,
+      "loss": 0.4746,
+      "step": 405
+    },
+    {
+      "epoch": 9.0,
+      "eval_combined_score": 0.3784420407842945,
+      "eval_loss": 2.554633855819702,
+      "eval_pearson": 0.38737881017296705,
+      "eval_runtime": 2.4989,
+      "eval_samples_per_second": 600.261,
+      "eval_spearmanr": 0.36950527139562195,
+      "eval_steps_per_second": 4.802,
+      "step": 405
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.4113,
+      "step": 450
+    },
+    {
+      "epoch": 10.0,
+      "eval_combined_score": 0.4075329782364955,
+      "eval_loss": 2.207718849182129,
+      "eval_pearson": 0.41124678689635685,
+      "eval_runtime": 2.501,
+      "eval_samples_per_second": 599.772,
+      "eval_spearmanr": 0.40381916957663416,
+      "eval_steps_per_second": 4.798,
+      "step": 450
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 3.1200000000000006e-05,
+      "loss": 0.3585,
+      "step": 495
+    },
+    {
+      "epoch": 11.0,
+      "eval_combined_score": 0.40338639765431117,
+      "eval_loss": 2.2846267223358154,
+      "eval_pearson": 0.40961605884447455,
+      "eval_runtime": 2.5166,
+      "eval_samples_per_second": 596.048,
+      "eval_spearmanr": 0.39715673646414784,
+      "eval_steps_per_second": 4.768,
+      "step": 495
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 3.0400000000000004e-05,
+      "loss": 0.3288,
+      "step": 540
+    },
+    {
+      "epoch": 12.0,
+      "eval_combined_score": 0.39296875304027296,
+      "eval_loss": 2.415489912033081,
+      "eval_pearson": 0.4011713493012907,
+      "eval_runtime": 2.4821,
+      "eval_samples_per_second": 604.328,
+      "eval_spearmanr": 0.3847661567792552,
+      "eval_steps_per_second": 4.835,
+      "step": 540
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 2.96e-05,
+      "loss": 0.2745,
+      "step": 585
+    },
+    {
+      "epoch": 13.0,
+      "eval_combined_score": 0.39641281291560126,
+      "eval_loss": 2.363490581512451,
+      "eval_pearson": 0.40043181134281525,
+      "eval_runtime": 2.473,
+      "eval_samples_per_second": 606.539,
+      "eval_spearmanr": 0.3923938144883873,
+      "eval_steps_per_second": 4.852,
+      "step": 585
+    },
+    {
+      "epoch": 14.0,
+      "learning_rate": 2.8800000000000002e-05,
+      "loss": 0.2579,
+      "step": 630
+    },
+    {
+      "epoch": 14.0,
+      "eval_combined_score": 0.4106013077457039,
+      "eval_loss": 2.0734338760375977,
+      "eval_pearson": 0.41838647449624344,
+      "eval_runtime": 2.493,
+      "eval_samples_per_second": 601.697,
+      "eval_spearmanr": 0.4028161409951644,
+      "eval_steps_per_second": 4.814,
+      "step": 630
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 2.8e-05,
+      "loss": 0.2309,
+      "step": 675
+    },
+    {
+      "epoch": 15.0,
+      "eval_combined_score": 0.4098630195139652,
+      "eval_loss": 2.3461852073669434,
+      "eval_pearson": 0.4170909773634089,
+      "eval_runtime": 2.4978,
+      "eval_samples_per_second": 600.52,
+      "eval_spearmanr": 0.40263506166452157,
+      "eval_steps_per_second": 4.804,
+      "step": 675
+    },
+    {
+      "epoch": 16.0,
+      "learning_rate": 2.7200000000000004e-05,
+      "loss": 0.2037,
+      "step": 720
+    },
+    {
+      "epoch": 16.0,
+      "eval_combined_score": 0.415740299608016,
+      "eval_loss": 2.259819269180298,
+      "eval_pearson": 0.4224942480806353,
+      "eval_runtime": 2.4898,
+      "eval_samples_per_second": 602.46,
+      "eval_spearmanr": 0.4089863511353967,
+      "eval_steps_per_second": 4.82,
+      "step": 720
+    },
+    {
+      "epoch": 17.0,
+      "learning_rate": 2.6400000000000005e-05,
+      "loss": 0.1806,
+      "step": 765
+    },
+    {
+      "epoch": 17.0,
+      "eval_combined_score": 0.4016026713274717,
+      "eval_loss": 2.2458109855651855,
+      "eval_pearson": 0.4116281690097336,
+      "eval_runtime": 2.4829,
+      "eval_samples_per_second": 604.129,
+      "eval_spearmanr": 0.3915771736452098,
+      "eval_steps_per_second": 4.833,
+      "step": 765
+    },
+    {
+      "epoch": 18.0,
+      "learning_rate": 2.5600000000000002e-05,
+      "loss": 0.1785,
+      "step": 810
+    },
+    {
+      "epoch": 18.0,
+      "eval_combined_score": 0.3995583326979918,
+      "eval_loss": 2.3296127319335938,
+      "eval_pearson": 0.4088303634529265,
+      "eval_runtime": 2.4855,
+      "eval_samples_per_second": 603.511,
+      "eval_spearmanr": 0.3902863019430571,
+      "eval_steps_per_second": 4.828,
+      "step": 810
+    },
+    {
+      "epoch": 19.0,
+      "learning_rate": 2.4800000000000003e-05,
+      "loss": 0.1582,
+      "step": 855
+    },
+    {
+      "epoch": 19.0,
+      "eval_combined_score": 0.39505603889568086,
+      "eval_loss": 2.3369128704071045,
+      "eval_pearson": 0.40326860099385287,
+      "eval_runtime": 2.4824,
+      "eval_samples_per_second": 604.243,
+      "eval_spearmanr": 0.3868434767975089,
+      "eval_steps_per_second": 4.834,
+      "step": 855
+    },
+    {
+      "epoch": 19.0,
+      "step": 855,
+      "total_flos": 1.5984645283774464e+16,
+      "train_loss": 0.7103834141067594,
+      "train_runtime": 695.8903,
+      "train_samples_per_second": 413.068,
+      "train_steps_per_second": 3.233
+    }
+  ],
+  "max_steps": 2250,
+  "num_train_epochs": 50,
+  "total_flos": 1.5984645283774464e+16,
+  "trial_name": null,
+  "trial_params": null
+}