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: 0.36816720257234725
+      value: 0.6318327974276527
     - name: F1
       type: f1
-      value: 0.5381903642773208
+      value: 0.0
 ---
 
 <!-- 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_pretrain_qqp
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new) on the GLUE QQP dataset.
 It achieves the following results on the evaluation set:
-- Loss: 3.0633
-- Accuracy: 0.3682
-- F1: 0.5382
-- Combined Score: 0.4532
+- Loss: 0.6569
+- Accuracy: 0.6318
+- F1: 0.0
+- Combined Score: 0.3159
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 11.0,
+    "eval_accuracy": 0.6318327974276527,
+    "eval_combined_score": 0.3159163987138264,
+    "eval_f1": 0.0,
+    "eval_loss": 0.6569251418113708,
+    "eval_runtime": 69.3881,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 582.665,
+    "eval_steps_per_second": 4.554,
+    "train_loss": 23.176317952643103,
+    "train_runtime": 20740.4239,
+    "train_samples": 363846,
+    "train_samples_per_second": 877.142,
+    "train_steps_per_second": 6.854
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 11.0,
+    "eval_accuracy": 0.6318327974276527,
+    "eval_combined_score": 0.3159163987138264,
+    "eval_f1": 0.0,
+    "eval_loss": 0.6569251418113708,
+    "eval_runtime": 69.3881,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 582.665,
+    "eval_steps_per_second": 4.554
+}

logs/events.out.tfevents.1685553025.serv-3318.2595329.14

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28abfb848be16226a5ea4dbe62465f6b9e0d169bd74a9c1d7d5be45ba04ac199
+size 475

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 11.0,
+    "train_loss": 23.176317952643103,
+    "train_runtime": 20740.4239,
+    "train_samples": 363846,
+    "train_samples_per_second": 877.142,
+    "train_steps_per_second": 6.854
+}

trainer_state.json

@@ -0,0 +1,212 @@
+{
+  "best_metric": 0.6569251418113708,
+  "best_model_checkpoint": "hBERTv2_new_pretrain_qqp/checkpoint-17058",
+  "epoch": 11.0,
+  "global_step": 31273,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.00049,
+      "loss": 130.0621,
+      "step": 2843
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.36816720257234725,
+      "eval_combined_score": 0.45317878342483403,
+      "eval_f1": 0.5381903642773208,
+      "eval_loss": 125.09819793701172,
+      "eval_runtime": 70.5346,
+      "eval_samples_per_second": 573.194,
+      "eval_steps_per_second": 4.48,
+      "step": 2843
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 0.00048,
+      "loss": 63.3796,
+      "step": 5686
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.36816720257234725,
+      "eval_combined_score": 0.45317878342483403,
+      "eval_f1": 0.5381903642773208,
+      "eval_loss": 0.8078107833862305,
+      "eval_runtime": 70.4435,
+      "eval_samples_per_second": 573.935,
+      "eval_steps_per_second": 4.486,
+      "step": 5686
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 0.00047,
+      "loss": 0.7936,
+      "step": 8529
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.36816720257234725,
+      "eval_combined_score": 0.45317878342483403,
+      "eval_f1": 0.5381903642773208,
+      "eval_loss": 2.612349271774292,
+      "eval_runtime": 70.4376,
+      "eval_samples_per_second": 573.983,
+      "eval_steps_per_second": 4.486,
+      "step": 8529
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 0.00046,
+      "loss": 27.5934,
+      "step": 11372
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.36816720257234725,
+      "eval_combined_score": 0.45317878342483403,
+      "eval_f1": 0.5381903642773208,
+      "eval_loss": 25.9051456451416,
+      "eval_runtime": 69.7747,
+      "eval_samples_per_second": 579.436,
+      "eval_steps_per_second": 4.529,
+      "step": 11372
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 0.00045000000000000004,
+      "loss": 20.3073,
+      "step": 14215
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.36816720257234725,
+      "eval_combined_score": 0.45317878342483403,
+      "eval_f1": 0.5381903642773208,
+      "eval_loss": 16.190715789794922,
+      "eval_runtime": 69.466,
+      "eval_samples_per_second": 582.011,
+      "eval_steps_per_second": 4.549,
+      "step": 14215
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 0.00044,
+      "loss": 7.6381,
+      "step": 17058
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6318327974276527,
+      "eval_combined_score": 0.3159163987138264,
+      "eval_f1": 0.0,
+      "eval_loss": 0.6569251418113708,
+      "eval_runtime": 69.8766,
+      "eval_samples_per_second": 578.592,
+      "eval_steps_per_second": 4.522,
+      "step": 17058
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 0.00043,
+      "loss": 0.6675,
+      "step": 19901
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6318327974276527,
+      "eval_combined_score": 0.3159163987138264,
+      "eval_f1": 0.0,
+      "eval_loss": 0.6590132713317871,
+      "eval_runtime": 70.184,
+      "eval_samples_per_second": 576.057,
+      "eval_steps_per_second": 4.502,
+      "step": 19901
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 0.00042,
+      "loss": 0.6629,
+      "step": 22744
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6318327974276527,
+      "eval_combined_score": 0.3159163987138264,
+      "eval_f1": 0.0,
+      "eval_loss": 0.6588091850280762,
+      "eval_runtime": 69.9082,
+      "eval_samples_per_second": 578.33,
+      "eval_steps_per_second": 4.52,
+      "step": 22744
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 0.00041,
+      "loss": 0.6629,
+      "step": 25587
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.6318327974276527,
+      "eval_combined_score": 0.3159163987138264,
+      "eval_f1": 0.0,
+      "eval_loss": 0.6590038537979126,
+      "eval_runtime": 70.2012,
+      "eval_samples_per_second": 575.916,
+      "eval_steps_per_second": 4.501,
+      "step": 25587
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 0.0004,
+      "loss": 0.6713,
+      "step": 28430
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.6318327974276527,
+      "eval_combined_score": 0.3159163987138264,
+      "eval_f1": 0.0,
+      "eval_loss": 0.6693503856658936,
+      "eval_runtime": 69.9698,
+      "eval_samples_per_second": 577.821,
+      "eval_steps_per_second": 4.516,
+      "step": 28430
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 0.00039000000000000005,
+      "loss": 2.5009,
+      "step": 31273
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.36816720257234725,
+      "eval_combined_score": 0.45317878342483403,
+      "eval_f1": 0.5381903642773208,
+      "eval_loss": 3.0633163452148438,
+      "eval_runtime": 69.917,
+      "eval_samples_per_second": 578.257,
+      "eval_steps_per_second": 4.52,
+      "step": 31273
+    },
+    {
+      "epoch": 11.0,
+      "step": 31273,
+      "total_flos": 5.856941202410045e+17,
+      "train_loss": 23.176317952643103,
+      "train_runtime": 20740.4239,
+      "train_samples_per_second": 877.142,
+      "train_steps_per_second": 6.854
+    }
+  ],
+  "max_steps": 142150,
+  "num_train_epochs": 50,
+  "total_flos": 5.856941202410045e+17,
+  "trial_name": null,
+  "trial_params": null
+}