End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 license: apache-2.0
 tags:
 - generated_from_trainer
@@ -14,7 +16,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE QQP
       type: glue
       config: qqp
       split: validation
@@ -22,10 +24,10 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.8579025476131585
+      value: 0.8510017313875835
     - name: F1
       type: f1
-      value: 0.8181328943619613
+      value: 0.799640790261425
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -33,12 +35,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # sa_BERT_48_qqp
 
-This model is a fine-tuned version of [gokuls/bert_base_48](https://huggingface.co/gokuls/bert_base_48) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_base_48](https://huggingface.co/gokuls/bert_base_48) on the GLUE QQP dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5063
-- Accuracy: 0.8579
-- F1: 0.8181
-- Combined Score: 0.8380
+- Loss: 0.3425
+- Accuracy: 0.8510
+- F1: 0.7996
+- Combined Score: 0.8253
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.8510017313875835,
+    "eval_combined_score": 0.8253212608245042,
+    "eval_f1": 0.799640790261425,
+    "eval_loss": 0.34253132343292236,
+    "eval_runtime": 84.7143,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 477.251,
+    "eval_steps_per_second": 4.981,
+    "train_loss": 0.24057756895378754,
+    "train_runtime": 19188.7559,
+    "train_samples": 363846,
+    "train_samples_per_second": 948.071,
+    "train_steps_per_second": 9.878
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.8510017313875835,
+    "eval_combined_score": 0.8253212608245042,
+    "eval_f1": 0.799640790261425,
+    "eval_loss": 0.34253132343292236,
+    "eval_runtime": 84.7143,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 477.251,
+    "eval_steps_per_second": 4.981
+}

logs/events.out.tfevents.1686890952.serv-3318.1430967.7

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9122623807bd90d18669d109ab57a6071c585d9472105ef7d388d79e67de5a16
+size 475

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 8.0,
+    "train_loss": 0.24057756895378754,
+    "train_runtime": 19188.7559,
+    "train_samples": 363846,
+    "train_samples_per_second": 948.071,
+    "train_steps_per_second": 9.878
+}

trainer_state.json

@@ -0,0 +1,161 @@
+{
+  "best_metric": 0.34253132343292236,
+  "best_model_checkpoint": "sa_BERT_48_qqp/checkpoint-11373",
+  "epoch": 8.0,
+  "global_step": 30328,
+  "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": 0.4679,
+      "step": 3791
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.8222359633935197,
+      "eval_combined_score": 0.796364668588026,
+      "eval_f1": 0.7704933737825324,
+      "eval_loss": 0.3794594705104828,
+      "eval_runtime": 85.3071,
+      "eval_samples_per_second": 473.935,
+      "eval_steps_per_second": 4.947,
+      "step": 3791
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.3469,
+      "step": 7582
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8446697996537225,
+      "eval_combined_score": 0.8204728353358471,
+      "eval_f1": 0.7962758710179718,
+      "eval_loss": 0.35802027583122253,
+      "eval_runtime": 85.9214,
+      "eval_samples_per_second": 470.546,
+      "eval_steps_per_second": 4.911,
+      "step": 7582
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.76e-05,
+      "loss": 0.2868,
+      "step": 11373
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.8510017313875835,
+      "eval_combined_score": 0.8253212608245042,
+      "eval_f1": 0.799640790261425,
+      "eval_loss": 0.34253132343292236,
+      "eval_runtime": 85.8338,
+      "eval_samples_per_second": 471.026,
+      "eval_steps_per_second": 4.916,
+      "step": 11373
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.2372,
+      "step": 15164
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8560969577046748,
+      "eval_combined_score": 0.8354877050200858,
+      "eval_f1": 0.8148784523354969,
+      "eval_loss": 0.37059277296066284,
+      "eval_runtime": 85.9472,
+      "eval_samples_per_second": 470.405,
+      "eval_steps_per_second": 4.91,
+      "step": 15164
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.1938,
+      "step": 18955
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.8624783576552065,
+      "eval_combined_score": 0.8410731175925228,
+      "eval_f1": 0.819667877529839,
+      "eval_loss": 0.36789774894714355,
+      "eval_runtime": 85.3085,
+      "eval_samples_per_second": 473.927,
+      "eval_steps_per_second": 4.947,
+      "step": 18955
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 3.52e-05,
+      "loss": 0.1567,
+      "step": 22746
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.8639376700469948,
+      "eval_combined_score": 0.8426697862913135,
+      "eval_f1": 0.8214019025356321,
+      "eval_loss": 0.42459815740585327,
+      "eval_runtime": 85.3092,
+      "eval_samples_per_second": 473.923,
+      "eval_steps_per_second": 4.947,
+      "step": 22746
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 3.44e-05,
+      "loss": 0.1294,
+      "step": 26537
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.8585456344298789,
+      "eval_combined_score": 0.8387393361321469,
+      "eval_f1": 0.8189330378344151,
+      "eval_loss": 0.40472903847694397,
+      "eval_runtime": 85.2488,
+      "eval_samples_per_second": 474.259,
+      "eval_steps_per_second": 4.95,
+      "step": 26537
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 0.1059,
+      "step": 30328
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.8579025476131585,
+      "eval_combined_score": 0.8380177209875599,
+      "eval_f1": 0.8181328943619613,
+      "eval_loss": 0.5063081383705139,
+      "eval_runtime": 85.299,
+      "eval_samples_per_second": 473.98,
+      "eval_steps_per_second": 4.947,
+      "step": 30328
+    },
+    {
+      "epoch": 8.0,
+      "step": 30328,
+      "total_flos": 3.8292763616005325e+17,
+      "train_loss": 0.24057756895378754,
+      "train_runtime": 19188.7559,
+      "train_samples_per_second": 948.071,
+      "train_steps_per_second": 9.878
+    }
+  ],
+  "max_steps": 189550,
+  "num_train_epochs": 50,
+  "total_flos": 3.8292763616005325e+17,
+  "trial_name": null,
+  "trial_params": null
+}