End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 base_model: gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_48
 tags:
 - generated_from_trainer
@@ -14,7 +16,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE MRPC
       type: glue
       config: mrpc
       split: validation
@@ -22,10 +24,10 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.6102941176470589
+      value: 0.7181372549019608
     - name: F1
       type: f1
-      value: 0.7145421903052065
+      value: 0.8099173553719009
 ---
 
 <!-- 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. -->
 
 # hBERTv1_new_pretrain_w_init_48_ver2_mrpc
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_48](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_48) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_48](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_48) on the GLUE MRPC dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.2159
-- Accuracy: 0.6103
-- F1: 0.7145
-- Combined Score: 0.6624
+- Loss: 0.5916
+- Accuracy: 0.7181
+- F1: 0.8099
+- Combined Score: 0.7640
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.7181372549019608,
+    "eval_combined_score": 0.7640273051369308,
+    "eval_f1": 0.8099173553719009,
+    "eval_loss": 0.5916191339492798,
+    "eval_runtime": 2.1093,
+    "eval_samples": 408,
+    "eval_samples_per_second": 193.426,
+    "eval_steps_per_second": 3.319,
+    "train_loss": 0.4669204134071989,
+    "train_runtime": 443.7082,
+    "train_samples": 3668,
+    "train_samples_per_second": 124.0,
+    "train_steps_per_second": 1.961
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.7181372549019608,
+    "eval_combined_score": 0.7640273051369308,
+    "eval_f1": 0.8099173553719009,
+    "eval_loss": 0.5916191339492798,
+    "eval_runtime": 2.1093,
+    "eval_samples": 408,
+    "eval_samples_per_second": 193.426,
+    "eval_steps_per_second": 3.319
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 7.0,
+    "train_loss": 0.4669204134071989,
+    "train_runtime": 443.7082,
+    "train_samples": 3668,
+    "train_samples_per_second": 124.0,
+    "train_steps_per_second": 1.961
+}

trainer_state.json

@@ -0,0 +1,147 @@
+{
+  "best_metric": 0.5916191339492798,
+  "best_model_checkpoint": "hBERTv1_new_pretrain_w_init_48_ver2_mrpc/checkpoint-116",
+  "epoch": 7.0,
+  "eval_steps": 500,
+  "global_step": 406,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.733333333333334e-05,
+      "loss": 0.6666,
+      "step": 58
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6911764705882353,
+      "eval_combined_score": 0.7459046909903202,
+      "eval_f1": 0.800632911392405,
+      "eval_loss": 0.6274183392524719,
+      "eval_runtime": 2.152,
+      "eval_samples_per_second": 189.587,
+      "eval_steps_per_second": 3.253,
+      "step": 58
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.466666666666667e-05,
+      "loss": 0.6239,
+      "step": 116
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.7181372549019608,
+      "eval_combined_score": 0.7640273051369308,
+      "eval_f1": 0.8099173553719009,
+      "eval_loss": 0.5916191339492798,
+      "eval_runtime": 2.1718,
+      "eval_samples_per_second": 187.865,
+      "eval_steps_per_second": 3.223,
+      "step": 116
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.5981,
+      "step": 174
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6225490196078431,
+      "eval_combined_score": 0.6614690623331044,
+      "eval_f1": 0.7003891050583657,
+      "eval_loss": 0.6531738638877869,
+      "eval_runtime": 2.173,
+      "eval_samples_per_second": 187.761,
+      "eval_steps_per_second": 3.221,
+      "step": 174
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 2.9333333333333333e-05,
+      "loss": 0.518,
+      "step": 232
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7107843137254902,
+      "eval_combined_score": 0.7583526831785345,
+      "eval_f1": 0.8059210526315789,
+      "eval_loss": 0.6250593662261963,
+      "eval_runtime": 2.1697,
+      "eval_samples_per_second": 188.047,
+      "eval_steps_per_second": 3.226,
+      "step": 232
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 2.6666666666666667e-05,
+      "loss": 0.3848,
+      "step": 290
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6813725490196079,
+      "eval_combined_score": 0.7341288974606237,
+      "eval_f1": 0.7868852459016394,
+      "eval_loss": 0.7553436160087585,
+      "eval_runtime": 2.1704,
+      "eval_samples_per_second": 187.987,
+      "eval_steps_per_second": 3.225,
+      "step": 290
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 2.4e-05,
+      "loss": 0.2708,
+      "step": 348
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7416007227152136,
+      "eval_f1": 0.7993779160186625,
+      "eval_loss": 1.069616436958313,
+      "eval_runtime": 2.1733,
+      "eval_samples_per_second": 187.73,
+      "eval_steps_per_second": 3.221,
+      "step": 348
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 2.1333333333333335e-05,
+      "loss": 0.2062,
+      "step": 406
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6102941176470589,
+      "eval_combined_score": 0.6624181539761327,
+      "eval_f1": 0.7145421903052065,
+      "eval_loss": 1.2159368991851807,
+      "eval_runtime": 2.1699,
+      "eval_samples_per_second": 188.026,
+      "eval_steps_per_second": 3.226,
+      "step": 406
+    },
+    {
+      "epoch": 7.0,
+      "step": 406,
+      "total_flos": 3802532820287488.0,
+      "train_loss": 0.4669204134071989,
+      "train_runtime": 443.7082,
+      "train_samples_per_second": 124.0,
+      "train_steps_per_second": 1.961
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 870,
+  "num_train_epochs": 15,
+  "save_steps": 500,
+  "total_flos": 3802532820287488.0,
+  "trial_name": null,
+  "trial_params": null
+}