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: 0.6318327974276527
+      value: 0.6823893148651992
     - name: F1
       type: f1
-      value: 0.0
+      value: 0.4704523897892697
 ---
 
 <!-- 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_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: 0.6774
-- Accuracy: 0.6318
-- F1: 0.0
-- Combined Score: 0.3159
+- Loss: 0.5939
+- Accuracy: 0.6824
+- F1: 0.4705
+- Combined Score: 0.5764
 
 ## Model description
 

all_results.json

@@ -1,16 +1,16 @@
 {
-    "epoch": 6.0,
-    "eval_accuracy": 0.6318327974276527,
-    "eval_combined_score": 0.3159163987138264,
-    "eval_f1": 0.0,
-    "eval_loss": NaN,
-    "eval_runtime": 63.0211,
+    "epoch": 9.0,
+    "eval_accuracy": 0.6823893148651992,
+    "eval_combined_score": 0.5764208523272344,
+    "eval_f1": 0.4704523897892697,
+    "eval_loss": 0.5939152240753174,
+    "eval_runtime": 75.5775,
     "eval_samples": 40430,
-    "eval_samples_per_second": 641.531,
-    "eval_steps_per_second": 5.014,
-    "train_loss": 0.12732445977052556,
-    "train_runtime": 9701.4674,
+    "eval_samples_per_second": 534.948,
+    "eval_steps_per_second": 4.181,
+    "train_loss": 0.6200644320571725,
+    "train_runtime": 19316.7897,
     "train_samples": 363846,
-    "train_samples_per_second": 1875.211,
-    "train_steps_per_second": 14.652
+    "train_samples_per_second": 941.787,
+    "train_steps_per_second": 7.359
 }

eval_results.json

@@ -1,11 +1,11 @@
 {
-    "epoch": 6.0,
-    "eval_accuracy": 0.6318327974276527,
-    "eval_combined_score": 0.3159163987138264,
-    "eval_f1": 0.0,
-    "eval_loss": NaN,
-    "eval_runtime": 63.0211,
+    "epoch": 9.0,
+    "eval_accuracy": 0.6823893148651992,
+    "eval_combined_score": 0.5764208523272344,
+    "eval_f1": 0.4704523897892697,
+    "eval_loss": 0.5939152240753174,
+    "eval_runtime": 75.5775,
     "eval_samples": 40430,
-    "eval_samples_per_second": 641.531,
-    "eval_steps_per_second": 5.014
+    "eval_samples_per_second": 534.948,
+    "eval_steps_per_second": 4.181
 }

logs/events.out.tfevents.1686772856.serv-9219.1715196.7

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:536b149a02c0c792a4539680891394a629f32294bd12b33c1a8f8ae402d0c9d4
+size 475

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 6.0,
-    "train_loss": 0.12732445977052556,
-    "train_runtime": 9701.4674,
+    "epoch": 9.0,
+    "train_loss": 0.6200644320571725,
+    "train_runtime": 19316.7897,
     "train_samples": 363846,
-    "train_samples_per_second": 1875.211,
-    "train_steps_per_second": 14.652
+    "train_samples_per_second": 941.787,
+    "train_steps_per_second": 7.359
 }

trainer_state.json

@@ -1,127 +1,178 @@
 {
-  "best_metric": NaN,
-  "best_model_checkpoint": "add_BERT_no_pretrain_qqp/checkpoint-2843",
-  "epoch": 6.0,
-  "global_step": 17058,
+  "best_metric": 0.5939152240753174,
+  "best_model_checkpoint": "add_BERT_no_pretrain_qqp/checkpoint-11372",
+  "epoch": 9.0,
+  "global_step": 25587,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.0,
-      "learning_rate": 0.000490583890256771,
-      "loss": 0.7639,
+      "learning_rate": 3.9200000000000004e-05,
+      "loss": 0.657,
       "step": 2843
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6318327974276527,
-      "eval_combined_score": 0.3159163987138264,
-      "eval_f1": 0.0,
-      "eval_loss": NaN,
-      "eval_runtime": 62.9642,
-      "eval_samples_per_second": 642.111,
-      "eval_steps_per_second": 5.019,
+      "eval_accuracy": 0.6490477368290873,
+      "eval_combined_score": 0.4049047757310398,
+      "eval_f1": 0.16076181463299224,
+      "eval_loss": 0.6437851786613464,
+      "eval_runtime": 75.6371,
+      "eval_samples_per_second": 534.526,
+      "eval_steps_per_second": 4.178,
       "step": 2843
     },
     {
       "epoch": 2.0,
-      "learning_rate": 0.000480583890256771,
-      "loss": 0.0,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.6273,
       "step": 5686
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.6318327974276527,
-      "eval_combined_score": 0.3159163987138264,
-      "eval_f1": 0.0,
-      "eval_loss": NaN,
-      "eval_runtime": 62.8999,
-      "eval_samples_per_second": 642.767,
-      "eval_steps_per_second": 5.024,
+      "eval_accuracy": 0.6442987880286916,
+      "eval_combined_score": 0.4180752158744188,
+      "eval_f1": 0.1918516437201461,
+      "eval_loss": 0.6302036643028259,
+      "eval_runtime": 75.8685,
+      "eval_samples_per_second": 532.896,
+      "eval_steps_per_second": 4.165,
       "step": 5686
     },
     {
       "epoch": 3.0,
-      "learning_rate": 0.00047058389025677103,
-      "loss": 0.0,
+      "learning_rate": 3.76e-05,
+      "loss": 0.6273,
       "step": 8529
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.6318327974276527,
-      "eval_combined_score": 0.3159163987138264,
-      "eval_f1": 0.0,
-      "eval_loss": NaN,
-      "eval_runtime": 62.9032,
-      "eval_samples_per_second": 642.733,
-      "eval_steps_per_second": 5.024,
+      "eval_accuracy": 0.6526589166460549,
+      "eval_combined_score": 0.5064287794766108,
+      "eval_f1": 0.36019864230716664,
+      "eval_loss": 0.6264548897743225,
+      "eval_runtime": 75.7943,
+      "eval_samples_per_second": 533.417,
+      "eval_steps_per_second": 4.169,
       "step": 8529
     },
     {
       "epoch": 4.0,
-      "learning_rate": 0.000460583890256771,
-      "loss": 0.0,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.6093,
       "step": 11372
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.6318327974276527,
-      "eval_combined_score": 0.3159163987138264,
-      "eval_f1": 0.0,
-      "eval_loss": NaN,
-      "eval_runtime": 63.0736,
-      "eval_samples_per_second": 640.997,
-      "eval_steps_per_second": 5.01,
+      "eval_accuracy": 0.6823893148651992,
+      "eval_combined_score": 0.5764208523272344,
+      "eval_f1": 0.4704523897892697,
+      "eval_loss": 0.5939152240753174,
+      "eval_runtime": 75.7663,
+      "eval_samples_per_second": 533.615,
+      "eval_steps_per_second": 4.171,
       "step": 11372
     },
     {
       "epoch": 5.0,
-      "learning_rate": 0.00045058389025677103,
-      "loss": 0.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.5932,
       "step": 14215
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.6318327974276527,
-      "eval_combined_score": 0.3159163987138264,
-      "eval_f1": 0.0,
-      "eval_loss": NaN,
-      "eval_runtime": 62.9276,
-      "eval_samples_per_second": 642.485,
-      "eval_steps_per_second": 5.022,
+      "eval_accuracy": 0.680187979223349,
+      "eval_combined_score": 0.5486151798731713,
+      "eval_f1": 0.41704238052299364,
+      "eval_loss": 0.5961658358573914,
+      "eval_runtime": 75.7813,
+      "eval_samples_per_second": 533.509,
+      "eval_steps_per_second": 4.17,
       "step": 14215
     },
     {
       "epoch": 6.0,
-      "learning_rate": 0.00044058389025677106,
-      "loss": 0.0,
+      "learning_rate": 3.52e-05,
+      "loss": 0.599,
       "step": 17058
     },
     {
       "epoch": 6.0,
+      "eval_accuracy": 0.675735839722978,
+      "eval_combined_score": 0.5776042813321474,
+      "eval_f1": 0.4794727229413166,
+      "eval_loss": 0.5981380343437195,
+      "eval_runtime": 75.8673,
+      "eval_samples_per_second": 532.904,
+      "eval_steps_per_second": 4.165,
+      "step": 17058
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 3.44e-05,
+      "loss": 0.6063,
+      "step": 19901
+    },
+    {
+      "epoch": 7.0,
       "eval_accuracy": 0.6318327974276527,
       "eval_combined_score": 0.3159163987138264,
       "eval_f1": 0.0,
-      "eval_loss": NaN,
-      "eval_runtime": 62.936,
-      "eval_samples_per_second": 642.398,
-      "eval_steps_per_second": 5.021,
-      "step": 17058
+      "eval_loss": 0.6510685086250305,
+      "eval_runtime": 75.7196,
+      "eval_samples_per_second": 533.944,
+      "eval_steps_per_second": 4.173,
+      "step": 19901
     },
     {
-      "epoch": 6.0,
-      "step": 17058,
-      "total_flos": 3.1179558180460954e+17,
-      "train_loss": 0.12732445977052556,
-      "train_runtime": 9701.4674,
-      "train_samples_per_second": 1875.211,
-      "train_steps_per_second": 14.652
+      "epoch": 8.0,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 0.6264,
+      "step": 22744
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6531783329210982,
+      "eval_combined_score": 0.4303083615769859,
+      "eval_f1": 0.2074383902328736,
+      "eval_loss": 0.6261195540428162,
+      "eval_runtime": 75.7764,
+      "eval_samples_per_second": 533.543,
+      "eval_steps_per_second": 4.17,
+      "step": 22744
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 3.28e-05,
+      "loss": 0.6348,
+      "step": 25587
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.6318327974276527,
+      "eval_combined_score": 0.3159163987138264,
+      "eval_f1": 0.0,
+      "eval_loss": 0.6774410605430603,
+      "eval_runtime": 75.9165,
+      "eval_samples_per_second": 532.559,
+      "eval_steps_per_second": 4.162,
+      "step": 25587
+    },
+    {
+      "epoch": 9.0,
+      "step": 25587,
+      "total_flos": 4.676933727069143e+17,
+      "train_loss": 0.6200644320571725,
+      "train_runtime": 19316.7897,
+      "train_samples_per_second": 941.787,
+      "train_steps_per_second": 7.359
     }
   ],
   "max_steps": 142150,
   "num_train_epochs": 50,
-  "total_flos": 3.1179558180460954e+17,
+  "total_flos": 4.676933727069143e+17,
   "trial_name": null,
   "trial_params": null
 }