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 QQP
       type: glue
       config: qqp
       split: validation
@@ -21,10 +23,10 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.8216423447934702
+      value: 0.7856047489488004
     - name: F1
       type: f1
-      value: 0.7376005239983989
+      value: 0.6930594900849859
 ---
 
 <!-- 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. -->
 
 # hBERTv2_new_no_pretrain_qqp
 
-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 QQP dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5113
-- Accuracy: 0.8216
-- F1: 0.7376
-- Combined Score: 0.7796
+- Loss: 0.4537
+- Accuracy: 0.7856
+- F1: 0.6931
+- Combined Score: 0.7393
 
 ## Model description
 

all_results.json

@@ -1,16 +1,16 @@
 {
-    "epoch": 10.0,
-    "eval_accuracy": 0.6318327974276527,
-    "eval_combined_score": 0.3159163987138264,
-    "eval_f1": 0.0,
-    "eval_loss": 0.6578601002693176,
-    "eval_runtime": 67.6032,
+    "epoch": 6.0,
+    "eval_accuracy": 0.7856047489488004,
+    "eval_combined_score": 0.7393321195168931,
+    "eval_f1": 0.6930594900849859,
+    "eval_loss": 0.45365220308303833,
+    "eval_runtime": 176.4334,
     "eval_samples": 40430,
-    "eval_samples_per_second": 598.048,
-    "eval_steps_per_second": 4.674,
-    "train_loss": 0.6595250953891686,
-    "train_runtime": 18380.3319,
+    "eval_samples_per_second": 229.152,
+    "eval_steps_per_second": 1.791,
+    "train_loss": 0.3369104450452592,
+    "train_runtime": 30144.3422,
     "train_samples": 363846,
-    "train_samples_per_second": 989.77,
-    "train_steps_per_second": 7.734
+    "train_samples_per_second": 603.506,
+    "train_steps_per_second": 4.716
 }

eval_results.json

@@ -1,11 +1,11 @@
 {
-    "epoch": 10.0,
-    "eval_accuracy": 0.6318327974276527,
-    "eval_combined_score": 0.3159163987138264,
-    "eval_f1": 0.0,
-    "eval_loss": 0.6578601002693176,
-    "eval_runtime": 67.6032,
+    "epoch": 6.0,
+    "eval_accuracy": 0.7856047489488004,
+    "eval_combined_score": 0.7393321195168931,
+    "eval_f1": 0.6930594900849859,
+    "eval_loss": 0.45365220308303833,
+    "eval_runtime": 176.4334,
     "eval_samples": 40430,
-    "eval_samples_per_second": 598.048,
-    "eval_steps_per_second": 4.674
+    "eval_samples_per_second": 229.152,
+    "eval_steps_per_second": 1.791
 }

logs/events.out.tfevents.1686791762.garda.2589545.7

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e5cff645d0b37cb27a62ba54bfde16b4230d22d9572e20b69e417c9102f2e91c
+size 475

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 10.0,
-    "train_loss": 0.6595250953891686,
-    "train_runtime": 18380.3319,
+    "epoch": 6.0,
+    "train_loss": 0.3369104450452592,
+    "train_runtime": 30144.3422,
     "train_samples": 363846,
-    "train_samples_per_second": 989.77,
-    "train_steps_per_second": 7.734
+    "train_samples_per_second": 603.506,
+    "train_steps_per_second": 4.716
 }

trainer_state.json

@@ -1,195 +1,127 @@
 {
-  "best_metric": 0.6578601002693176,
-  "best_model_checkpoint": "hBERTv2_new_no_pretrain_qqp/checkpoint-14215",
-  "epoch": 10.0,
-  "global_step": 28430,
+  "best_metric": 0.45365220308303833,
+  "best_model_checkpoint": "hBERTv2_new_no_pretrain_qqp/checkpoint-2843",
+  "epoch": 6.0,
+  "global_step": 17058,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.0,
-      "learning_rate": 0.0004900105522335561,
-      "loss": 0.6669,
+      "learning_rate": 3.9200000000000004e-05,
+      "loss": 0.5037,
       "step": 2843
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6318327974276527,
-      "eval_combined_score": 0.3159163987138264,
-      "eval_f1": 0.0,
-      "eval_loss": 0.6595353484153748,
-      "eval_runtime": 67.7356,
-      "eval_samples_per_second": 596.88,
-      "eval_steps_per_second": 4.665,
+      "eval_accuracy": 0.7856047489488004,
+      "eval_combined_score": 0.7393321195168931,
+      "eval_f1": 0.6930594900849859,
+      "eval_loss": 0.45365220308303833,
+      "eval_runtime": 177.1269,
+      "eval_samples_per_second": 228.254,
+      "eval_steps_per_second": 1.784,
       "step": 2843
     },
     {
       "epoch": 2.0,
-      "learning_rate": 0.0004800105522335561,
-      "loss": 0.6591,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.4066,
       "step": 5686
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.6318327974276527,
-      "eval_combined_score": 0.3159163987138264,
-      "eval_f1": 0.0,
-      "eval_loss": 0.6586501598358154,
-      "eval_runtime": 67.7263,
-      "eval_samples_per_second": 596.962,
-      "eval_steps_per_second": 4.666,
+      "eval_accuracy": 0.794607964382884,
+      "eval_combined_score": 0.7351924719618621,
+      "eval_f1": 0.6757769795408404,
+      "eval_loss": 0.45485788583755493,
+      "eval_runtime": 179.191,
+      "eval_samples_per_second": 225.625,
+      "eval_steps_per_second": 1.763,
       "step": 5686
     },
     {
       "epoch": 3.0,
-      "learning_rate": 0.0004700105522335561,
-      "loss": 0.6589,
+      "learning_rate": 3.76e-05,
+      "loss": 0.3367,
       "step": 8529
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.6318327974276527,
-      "eval_combined_score": 0.3159163987138264,
-      "eval_f1": 0.0,
-      "eval_loss": 0.6581756472587585,
-      "eval_runtime": 67.7959,
-      "eval_samples_per_second": 596.348,
-      "eval_steps_per_second": 4.661,
+      "eval_accuracy": 0.7950284442245857,
+      "eval_combined_score": 0.7300256413289143,
+      "eval_f1": 0.6650228384332431,
+      "eval_loss": 0.4630189836025238,
+      "eval_runtime": 177.5842,
+      "eval_samples_per_second": 227.667,
+      "eval_steps_per_second": 1.779,
       "step": 8529
     },
     {
       "epoch": 4.0,
-      "learning_rate": 0.0004600140696447415,
-      "loss": 0.6587,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.2876,
       "step": 11372
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.6318327974276527,
-      "eval_combined_score": 0.3159163987138264,
-      "eval_f1": 0.0,
-      "eval_loss": 0.6579899787902832,
-      "eval_runtime": 67.7187,
-      "eval_samples_per_second": 597.028,
-      "eval_steps_per_second": 4.666,
+      "eval_accuracy": 0.8180064308681672,
+      "eval_combined_score": 0.7889000553400642,
+      "eval_f1": 0.7597936798119612,
+      "eval_loss": 0.5278903841972351,
+      "eval_runtime": 177.7849,
+      "eval_samples_per_second": 227.41,
+      "eval_steps_per_second": 1.777,
       "step": 11372
     },
     {
       "epoch": 5.0,
-      "learning_rate": 0.0004500175870559268,
-      "loss": 0.6586,
+      "learning_rate": 3.6e-05,
+      "loss": 0.2498,
       "step": 14215
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.6318327974276527,
-      "eval_combined_score": 0.3159163987138264,
-      "eval_f1": 0.0,
-      "eval_loss": 0.6578601002693176,
-      "eval_runtime": 67.6584,
-      "eval_samples_per_second": 597.561,
-      "eval_steps_per_second": 4.671,
+      "eval_accuracy": 0.8216670789018056,
+      "eval_combined_score": 0.7933379394248289,
+      "eval_f1": 0.7650087999478522,
+      "eval_loss": 0.4857378900051117,
+      "eval_runtime": 178.1608,
+      "eval_samples_per_second": 226.93,
+      "eval_steps_per_second": 1.774,
       "step": 14215
     },
     {
       "epoch": 6.0,
-      "learning_rate": 0.0004400211044671122,
-      "loss": 0.6586,
+      "learning_rate": 3.52e-05,
+      "loss": 0.2371,
       "step": 17058
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.6318327974276527,
-      "eval_combined_score": 0.3159163987138264,
-      "eval_f1": 0.0,
-      "eval_loss": 0.6580154895782471,
-      "eval_runtime": 67.6879,
-      "eval_samples_per_second": 597.3,
-      "eval_steps_per_second": 4.668,
+      "eval_accuracy": 0.8216423447934702,
+      "eval_combined_score": 0.7796214343959346,
+      "eval_f1": 0.7376005239983989,
+      "eval_loss": 0.5113059282302856,
+      "eval_runtime": 177.6527,
+      "eval_samples_per_second": 227.579,
+      "eval_steps_per_second": 1.779,
       "step": 17058
     },
     {
-      "epoch": 7.0,
-      "learning_rate": 0.00043002110446711226,
-      "loss": 0.6586,
-      "step": 19901
-    },
-    {
-      "epoch": 7.0,
-      "eval_accuracy": 0.6318327974276527,
-      "eval_combined_score": 0.3159163987138264,
-      "eval_f1": 0.0,
-      "eval_loss": 0.6579923629760742,
-      "eval_runtime": 67.6844,
-      "eval_samples_per_second": 597.331,
-      "eval_steps_per_second": 4.669,
-      "step": 19901
-    },
-    {
-      "epoch": 8.0,
-      "learning_rate": 0.00042002813928948297,
-      "loss": 0.6586,
-      "step": 22744
-    },
-    {
-      "epoch": 8.0,
-      "eval_accuracy": 0.6318327974276527,
-      "eval_combined_score": 0.3159163987138264,
-      "eval_f1": 0.0,
-      "eval_loss": 0.6579098701477051,
-      "eval_runtime": 67.6644,
-      "eval_samples_per_second": 597.508,
-      "eval_steps_per_second": 4.67,
-      "step": 22744
-    },
-    {
-      "epoch": 9.0,
-      "learning_rate": 0.00041003165670066834,
-      "loss": 0.6586,
-      "step": 25587
-    },
-    {
-      "epoch": 9.0,
-      "eval_accuracy": 0.6318327974276527,
-      "eval_combined_score": 0.3159163987138264,
-      "eval_f1": 0.0,
-      "eval_loss": 0.6579644680023193,
-      "eval_runtime": 67.7204,
-      "eval_samples_per_second": 597.014,
-      "eval_steps_per_second": 4.666,
-      "step": 25587
-    },
-    {
-      "epoch": 10.0,
-      "learning_rate": 0.0004000351741118537,
-      "loss": 0.6586,
-      "step": 28430
-    },
-    {
-      "epoch": 10.0,
-      "eval_accuracy": 0.6318327974276527,
-      "eval_combined_score": 0.3159163987138264,
-      "eval_f1": 0.0,
-      "eval_loss": 0.6579644680023193,
-      "eval_runtime": 67.7641,
-      "eval_samples_per_second": 596.629,
-      "eval_steps_per_second": 4.663,
-      "step": 28430
-    },
-    {
-      "epoch": 10.0,
-      "step": 28430,
-      "total_flos": 5.3244920021909504e+17,
-      "train_loss": 0.6595250953891686,
-      "train_runtime": 18380.3319,
-      "train_samples_per_second": 989.77,
-      "train_steps_per_second": 7.734
+      "epoch": 6.0,
+      "step": 17058,
+      "total_flos": 3.19469520131457e+17,
+      "train_loss": 0.3369104450452592,
+      "train_runtime": 30144.3422,
+      "train_samples_per_second": 603.506,
+      "train_steps_per_second": 4.716
     }
   ],
   "max_steps": 142150,
   "num_train_epochs": 50,
-  "total_flos": 5.3244920021909504e+17,
+  "total_flos": 3.19469520131457e+17,
   "trial_name": null,
   "trial_params": null
 }