End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 tags:
 - generated_from_trainer
 datasets:
@@ -13,7 +15,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE MRPC
       type: glue
       config: mrpc
       split: validation
@@ -21,10 +23,10 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.6348039215686274
+      value: 0.696078431372549
     - name: F1
       type: f1
-      value: 0.7285974499089254
+      value: 0.7933333333333332
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -32,12 +34,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # add_BERT_no_pretrain_mrpc
 
-This model is a fine-tuned version of [](https://huggingface.co/) on the glue dataset.
+This model is a fine-tuned version of [](https://huggingface.co/) on the GLUE MRPC dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6322
-- Accuracy: 0.6348
-- F1: 0.7286
-- Combined Score: 0.6817
+- Loss: 0.5912
+- Accuracy: 0.6961
+- F1: 0.7933
+- Combined Score: 0.7447
 
 ## Model description
 

all_results.json

@@ -1,16 +1,16 @@
 {
-    "epoch": 19.0,
-    "eval_accuracy": 0.6838235294117647,
-    "eval_combined_score": 0.7480253018237863,
-    "eval_f1": 0.8122270742358079,
-    "eval_loss": 0.6239899396896362,
-    "eval_runtime": 0.6442,
+    "epoch": 12.0,
+    "eval_accuracy": 0.696078431372549,
+    "eval_combined_score": 0.7447058823529411,
+    "eval_f1": 0.7933333333333332,
+    "eval_loss": 0.5912045240402222,
+    "eval_runtime": 0.7853,
     "eval_samples": 408,
-    "eval_samples_per_second": 633.308,
-    "eval_steps_per_second": 6.209,
-    "train_loss": 0.6928667667340453,
-    "train_runtime": 412.0299,
+    "eval_samples_per_second": 519.57,
+    "eval_steps_per_second": 5.094,
+    "train_loss": 0.5847925564338421,
+    "train_runtime": 324.1281,
     "train_samples": 3668,
-    "train_samples_per_second": 445.113,
-    "train_steps_per_second": 3.519
+    "train_samples_per_second": 565.826,
+    "train_steps_per_second": 4.474
 }

eval_results.json

@@ -1,11 +1,11 @@
 {
-    "epoch": 19.0,
-    "eval_accuracy": 0.6838235294117647,
-    "eval_combined_score": 0.7480253018237863,
-    "eval_f1": 0.8122270742358079,
-    "eval_loss": 0.6239899396896362,
-    "eval_runtime": 0.6442,
+    "epoch": 12.0,
+    "eval_accuracy": 0.696078431372549,
+    "eval_combined_score": 0.7447058823529411,
+    "eval_f1": 0.7933333333333332,
+    "eval_loss": 0.5912045240402222,
+    "eval_runtime": 0.7853,
     "eval_samples": 408,
-    "eval_samples_per_second": 633.308,
-    "eval_steps_per_second": 6.209
+    "eval_samples_per_second": 519.57,
+    "eval_steps_per_second": 5.094
 }

logs/events.out.tfevents.1686747304.serv-9219.1715196.3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f1141fad4909ed2702e99f35425546d07b5eb9e0e5ac131f3ede6eead7b0cf5b
+size 467

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 19.0,
-    "train_loss": 0.6928667667340453,
-    "train_runtime": 412.0299,
+    "epoch": 12.0,
+    "train_loss": 0.5847925564338421,
+    "train_runtime": 324.1281,
     "train_samples": 3668,
-    "train_samples_per_second": 445.113,
-    "train_steps_per_second": 3.519
+    "train_samples_per_second": 565.826,
+    "train_steps_per_second": 4.474
 }

trainer_state.json

@@ -1,16 +1,16 @@
 {
-  "best_metric": 0.6239899396896362,
-  "best_model_checkpoint": "add_BERT_no_pretrain_mrpc/checkpoint-406",
-  "epoch": 19.0,
-  "global_step": 551,
+  "best_metric": 0.5912045240402222,
+  "best_model_checkpoint": "add_BERT_no_pretrain_mrpc/checkpoint-203",
+  "epoch": 12.0,
+  "global_step": 348,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.0,
-      "learning_rate": 0.0004910344827586207,
-      "loss": 1.154,
+      "learning_rate": 3.9200000000000004e-05,
+      "loss": 0.6854,
       "step": 29
     },
     {
@@ -18,16 +18,16 @@
       "eval_accuracy": 0.6838235294117647,
       "eval_combined_score": 0.7480253018237863,
       "eval_f1": 0.8122270742358079,
-      "eval_loss": 0.6856198906898499,
-      "eval_runtime": 0.6445,
-      "eval_samples_per_second": 633.061,
-      "eval_steps_per_second": 6.206,
+      "eval_loss": 0.6710845828056335,
+      "eval_runtime": 0.7677,
+      "eval_samples_per_second": 531.445,
+      "eval_steps_per_second": 5.21,
       "step": 29
     },
     {
       "epoch": 2.0,
-      "learning_rate": 0.0004810344827586207,
-      "loss": 0.6781,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.6496,
       "step": 58
     },
     {
@@ -35,16 +35,16 @@
       "eval_accuracy": 0.6838235294117647,
       "eval_combined_score": 0.7480253018237863,
       "eval_f1": 0.8122270742358079,
-      "eval_loss": 0.6609018445014954,
-      "eval_runtime": 0.634,
-      "eval_samples_per_second": 643.583,
-      "eval_steps_per_second": 6.31,
+      "eval_loss": 0.6802303791046143,
+      "eval_runtime": 0.7711,
+      "eval_samples_per_second": 529.088,
+      "eval_steps_per_second": 5.187,
       "step": 58
     },
     {
       "epoch": 3.0,
-      "learning_rate": 0.00047103448275862073,
-      "loss": 0.6458,
+      "learning_rate": 3.76e-05,
+      "loss": 0.648,
       "step": 87
     },
     {
@@ -52,297 +52,178 @@
       "eval_accuracy": 0.6838235294117647,
       "eval_combined_score": 0.7480253018237863,
       "eval_f1": 0.8122270742358079,
-      "eval_loss": 0.634788990020752,
-      "eval_runtime": 0.6427,
-      "eval_samples_per_second": 634.811,
-      "eval_steps_per_second": 6.224,
+      "eval_loss": 0.6246173977851868,
+      "eval_runtime": 0.7699,
+      "eval_samples_per_second": 529.951,
+      "eval_steps_per_second": 5.196,
       "step": 87
     },
     {
       "epoch": 4.0,
-      "learning_rate": 0.0004610344827586207,
-      "loss": 0.6395,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.6363,
       "step": 116
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.31862745098039214,
-      "eval_combined_score": 0.16288515406162463,
-      "eval_f1": 0.0071428571428571435,
-      "eval_loss": 19.635393142700195,
-      "eval_runtime": 0.6439,
-      "eval_samples_per_second": 633.614,
-      "eval_steps_per_second": 6.212,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6174120903015137,
+      "eval_runtime": 0.7694,
+      "eval_samples_per_second": 530.299,
+      "eval_steps_per_second": 5.199,
       "step": 116
     },
     {
       "epoch": 5.0,
-      "learning_rate": 0.0004510344827586207,
-      "loss": 1.1486,
+      "learning_rate": 3.6e-05,
+      "loss": 0.6049,
       "step": 145
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.6838235294117647,
-      "eval_combined_score": 0.7480253018237863,
-      "eval_f1": 0.8122270742358079,
-      "eval_loss": 0.6657487750053406,
-      "eval_runtime": 0.6436,
-      "eval_samples_per_second": 633.968,
-      "eval_steps_per_second": 6.215,
+      "eval_accuracy": 0.6593137254901961,
+      "eval_combined_score": 0.7026001899845862,
+      "eval_f1": 0.7458866544789762,
+      "eval_loss": 0.6176258325576782,
+      "eval_runtime": 0.7696,
+      "eval_samples_per_second": 530.143,
+      "eval_steps_per_second": 5.197,
       "step": 145
     },
     {
       "epoch": 6.0,
-      "learning_rate": 0.00044103448275862066,
-      "loss": 0.6446,
+      "learning_rate": 3.52e-05,
+      "loss": 0.5491,
       "step": 174
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.6838235294117647,
-      "eval_combined_score": 0.7480253018237863,
-      "eval_f1": 0.8122270742358079,
-      "eval_loss": 0.627680778503418,
-      "eval_runtime": 0.6427,
-      "eval_samples_per_second": 634.78,
-      "eval_steps_per_second": 6.223,
+      "eval_accuracy": 0.6813725490196079,
+      "eval_combined_score": 0.7381626152038103,
+      "eval_f1": 0.7949526813880127,
+      "eval_loss": 0.6037724018096924,
+      "eval_runtime": 0.7688,
+      "eval_samples_per_second": 530.695,
+      "eval_steps_per_second": 5.203,
       "step": 174
     },
     {
       "epoch": 7.0,
-      "learning_rate": 0.0004310344827586207,
-      "loss": 0.644,
+      "learning_rate": 3.44e-05,
+      "loss": 0.5601,
       "step": 203
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.6838235294117647,
-      "eval_combined_score": 0.7480253018237863,
-      "eval_f1": 0.8122270742358079,
-      "eval_loss": 0.6241939663887024,
-      "eval_runtime": 0.6436,
-      "eval_samples_per_second": 633.911,
-      "eval_steps_per_second": 6.215,
+      "eval_accuracy": 0.696078431372549,
+      "eval_combined_score": 0.7447058823529411,
+      "eval_f1": 0.7933333333333332,
+      "eval_loss": 0.5912045240402222,
+      "eval_runtime": 0.7735,
+      "eval_samples_per_second": 527.497,
+      "eval_steps_per_second": 5.172,
       "step": 203
     },
     {
       "epoch": 8.0,
-      "learning_rate": 0.0004210344827586207,
-      "loss": 0.6337,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 0.5505,
       "step": 232
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.6838235294117647,
-      "eval_combined_score": 0.7480253018237863,
-      "eval_f1": 0.8122270742358079,
-      "eval_loss": 0.6242448687553406,
-      "eval_runtime": 0.6458,
-      "eval_samples_per_second": 631.822,
-      "eval_steps_per_second": 6.194,
+      "eval_accuracy": 0.6715686274509803,
+      "eval_combined_score": 0.7248571614076094,
+      "eval_f1": 0.7781456953642384,
+      "eval_loss": 0.6346378922462463,
+      "eval_runtime": 0.772,
+      "eval_samples_per_second": 528.513,
+      "eval_steps_per_second": 5.182,
       "step": 232
     },
     {
       "epoch": 9.0,
-      "learning_rate": 0.0004110344827586207,
-      "loss": 0.6388,
+      "learning_rate": 3.28e-05,
+      "loss": 0.5327,
       "step": 261
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.6838235294117647,
-      "eval_combined_score": 0.7480253018237863,
-      "eval_f1": 0.8122270742358079,
-      "eval_loss": 0.6252579092979431,
-      "eval_runtime": 0.6457,
-      "eval_samples_per_second": 631.91,
-      "eval_steps_per_second": 6.195,
+      "eval_accuracy": 0.6544117647058824,
+      "eval_combined_score": 0.7037382816524158,
+      "eval_f1": 0.7530647985989491,
+      "eval_loss": 0.628277063369751,
+      "eval_runtime": 0.7697,
+      "eval_samples_per_second": 530.071,
+      "eval_steps_per_second": 5.197,
       "step": 261
     },
     {
       "epoch": 10.0,
-      "learning_rate": 0.0004010344827586207,
-      "loss": 0.634,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.529,
       "step": 290
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.6838235294117647,
-      "eval_combined_score": 0.7480253018237863,
-      "eval_f1": 0.8122270742358079,
-      "eval_loss": 0.6241927742958069,
-      "eval_runtime": 0.6462,
-      "eval_samples_per_second": 631.361,
-      "eval_steps_per_second": 6.19,
+      "eval_accuracy": 0.6519607843137255,
+      "eval_combined_score": 0.7044050496911094,
+      "eval_f1": 0.7568493150684933,
+      "eval_loss": 0.634149432182312,
+      "eval_runtime": 0.7681,
+      "eval_samples_per_second": 531.164,
+      "eval_steps_per_second": 5.207,
       "step": 290
     },
     {
       "epoch": 11.0,
-      "learning_rate": 0.0003910344827586207,
-      "loss": 0.6346,
+      "learning_rate": 3.1200000000000006e-05,
+      "loss": 0.5337,
       "step": 319
     },
     {
       "epoch": 11.0,
-      "eval_accuracy": 0.6838235294117647,
-      "eval_combined_score": 0.7480253018237863,
-      "eval_f1": 0.8122270742358079,
-      "eval_loss": 0.6264073848724365,
-      "eval_runtime": 0.6465,
-      "eval_samples_per_second": 631.046,
-      "eval_steps_per_second": 6.187,
+      "eval_accuracy": 0.6617647058823529,
+      "eval_combined_score": 0.7098297213622291,
+      "eval_f1": 0.7578947368421053,
+      "eval_loss": 0.6285053491592407,
+      "eval_runtime": 0.7705,
+      "eval_samples_per_second": 529.529,
+      "eval_steps_per_second": 5.191,
       "step": 319
     },
     {
       "epoch": 12.0,
-      "learning_rate": 0.00038103448275862066,
-      "loss": 0.6338,
+      "learning_rate": 3.0400000000000004e-05,
+      "loss": 0.5383,
       "step": 348
     },
     {
       "epoch": 12.0,
-      "eval_accuracy": 0.6838235294117647,
-      "eval_combined_score": 0.7480253018237863,
-      "eval_f1": 0.8122270742358079,
-      "eval_loss": 0.6273019909858704,
-      "eval_runtime": 0.6464,
-      "eval_samples_per_second": 631.183,
-      "eval_steps_per_second": 6.188,
+      "eval_accuracy": 0.6348039215686274,
+      "eval_combined_score": 0.6817006857387764,
+      "eval_f1": 0.7285974499089254,
+      "eval_loss": 0.632221519947052,
+      "eval_runtime": 0.7713,
+      "eval_samples_per_second": 528.973,
+      "eval_steps_per_second": 5.186,
       "step": 348
     },
     {
-      "epoch": 13.0,
-      "learning_rate": 0.0003710344827586207,
-      "loss": 0.6343,
-      "step": 377
-    },
-    {
-      "epoch": 13.0,
-      "eval_accuracy": 0.6838235294117647,
-      "eval_combined_score": 0.7480253018237863,
-      "eval_f1": 0.8122270742358079,
-      "eval_loss": 0.6262201070785522,
-      "eval_runtime": 0.6462,
-      "eval_samples_per_second": 631.42,
-      "eval_steps_per_second": 6.19,
-      "step": 377
-    },
-    {
-      "epoch": 14.0,
-      "learning_rate": 0.0003610344827586207,
-      "loss": 0.6339,
-      "step": 406
-    },
-    {
-      "epoch": 14.0,
-      "eval_accuracy": 0.6838235294117647,
-      "eval_combined_score": 0.7480253018237863,
-      "eval_f1": 0.8122270742358079,
-      "eval_loss": 0.6239899396896362,
-      "eval_runtime": 0.6481,
-      "eval_samples_per_second": 629.552,
-      "eval_steps_per_second": 6.172,
-      "step": 406
-    },
-    {
-      "epoch": 15.0,
-      "learning_rate": 0.0003510344827586207,
-      "loss": 0.635,
-      "step": 435
-    },
-    {
-      "epoch": 15.0,
-      "eval_accuracy": 0.6838235294117647,
-      "eval_combined_score": 0.7480253018237863,
-      "eval_f1": 0.8122270742358079,
-      "eval_loss": 0.6244243383407593,
-      "eval_runtime": 0.6472,
-      "eval_samples_per_second": 630.443,
-      "eval_steps_per_second": 6.181,
-      "step": 435
-    },
-    {
-      "epoch": 16.0,
-      "learning_rate": 0.0003410344827586207,
-      "loss": 0.6331,
-      "step": 464
-    },
-    {
-      "epoch": 16.0,
-      "eval_accuracy": 0.6838235294117647,
-      "eval_combined_score": 0.7480253018237863,
-      "eval_f1": 0.8122270742358079,
-      "eval_loss": 0.6240312457084656,
-      "eval_runtime": 0.6454,
-      "eval_samples_per_second": 632.149,
-      "eval_steps_per_second": 6.198,
-      "step": 464
-    },
-    {
-      "epoch": 17.0,
-      "learning_rate": 0.0003310344827586207,
-      "loss": 0.6328,
-      "step": 493
-    },
-    {
-      "epoch": 17.0,
-      "eval_accuracy": 0.6838235294117647,
-      "eval_combined_score": 0.7480253018237863,
-      "eval_f1": 0.8122270742358079,
-      "eval_loss": 0.6266641020774841,
-      "eval_runtime": 0.646,
-      "eval_samples_per_second": 631.583,
-      "eval_steps_per_second": 6.192,
-      "step": 493
-    },
-    {
-      "epoch": 18.0,
-      "learning_rate": 0.00032103448275862067,
-      "loss": 0.6338,
-      "step": 522
-    },
-    {
-      "epoch": 18.0,
-      "eval_accuracy": 0.6838235294117647,
-      "eval_combined_score": 0.7480253018237863,
-      "eval_f1": 0.8122270742358079,
-      "eval_loss": 0.6256989240646362,
-      "eval_runtime": 0.6466,
-      "eval_samples_per_second": 630.959,
-      "eval_steps_per_second": 6.186,
-      "step": 522
-    },
-    {
-      "epoch": 19.0,
-      "learning_rate": 0.0003110344827586207,
-      "loss": 0.6321,
-      "step": 551
-    },
-    {
-      "epoch": 19.0,
-      "eval_accuracy": 0.6838235294117647,
-      "eval_combined_score": 0.7480253018237863,
-      "eval_f1": 0.8122270742358079,
-      "eval_loss": 0.623990535736084,
-      "eval_runtime": 0.6488,
-      "eval_samples_per_second": 628.807,
-      "eval_steps_per_second": 6.165,
-      "step": 551
-    },
-    {
-      "epoch": 19.0,
-      "step": 551,
-      "total_flos": 9953687707320320.0,
-      "train_loss": 0.6928667667340453,
-      "train_runtime": 412.0299,
-      "train_samples_per_second": 445.113,
-      "train_steps_per_second": 3.519
+      "epoch": 12.0,
+      "step": 348,
+      "total_flos": 6286539604623360.0,
+      "train_loss": 0.5847925564338421,
+      "train_runtime": 324.1281,
+      "train_samples_per_second": 565.826,
+      "train_steps_per_second": 4.474
     }
   ],
   "max_steps": 1450,
   "num_train_epochs": 50,
-  "total_flos": 9953687707320320.0,
+  "total_flos": 6286539604623360.0,
   "trial_name": null,
   "trial_params": null
 }