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 MRPC
       type: glue
       config: mrpc
       split: validation
@@ -21,10 +23,10 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.6544117647058824
+      value: 0.7058823529411765
     - name: F1
       type: f1
-      value: 0.7477638640429337
+      value: 0.8192771084337349
 ---
 
 <!-- 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_w_init__mrpc
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new_wt_init](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new_wt_init) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new_wt_init](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new_wt_init) on the GLUE MRPC dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.8311
-- Accuracy: 0.6544
-- F1: 0.7478
-- Combined Score: 0.7011
+- Loss: 0.5908
+- Accuracy: 0.7059
+- F1: 0.8193
+- Combined Score: 0.7626
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 6.0,
+    "eval_accuracy": 0.7058823529411765,
+    "eval_combined_score": 0.7625797306874558,
+    "eval_f1": 0.8192771084337349,
+    "eval_loss": 0.5908466577529907,
+    "eval_runtime": 0.7084,
+    "eval_samples": 408,
+    "eval_samples_per_second": 575.986,
+    "eval_steps_per_second": 5.647,
+    "train_loss": 0.48217608736849377,
+    "train_runtime": 133.0395,
+    "train_samples": 3668,
+    "train_samples_per_second": 1378.538,
+    "train_steps_per_second": 10.899
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 6.0,
+    "eval_accuracy": 0.7058823529411765,
+    "eval_combined_score": 0.7625797306874558,
+    "eval_f1": 0.8192771084337349,
+    "eval_loss": 0.5908466577529907,
+    "eval_runtime": 0.7084,
+    "eval_samples": 408,
+    "eval_samples_per_second": 575.986,
+    "eval_steps_per_second": 5.647
+}

logs/events.out.tfevents.1686033539.serv-3317.3380349.8

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:299fe6e8337494245dd72fcbcbd95a59c508a02b6e5bd4c1cb15dae5d02e20b1
+size 467

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 6.0,
+    "train_loss": 0.48217608736849377,
+    "train_runtime": 133.0395,
+    "train_samples": 3668,
+    "train_samples_per_second": 1378.538,
+    "train_steps_per_second": 10.899
+}

trainer_state.json

@@ -0,0 +1,127 @@
+{
+  "best_metric": 0.5908466577529907,
+  "best_model_checkpoint": "hBERTv2_new_pretrain_w_init__mrpc/checkpoint-29",
+  "epoch": 6.0,
+  "global_step": 174,
+  "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": 0.6576,
+      "step": 29
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7058823529411765,
+      "eval_combined_score": 0.7625797306874558,
+      "eval_f1": 0.8192771084337349,
+      "eval_loss": 0.5908466577529907,
+      "eval_runtime": 0.7089,
+      "eval_samples_per_second": 575.548,
+      "eval_steps_per_second": 5.643,
+      "step": 29
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.6172,
+      "step": 58
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6495098039215687,
+      "eval_combined_score": 0.6963886542049496,
+      "eval_f1": 0.7432675044883303,
+      "eval_loss": 0.6228410601615906,
+      "eval_runtime": 0.7086,
+      "eval_samples_per_second": 575.747,
+      "eval_steps_per_second": 5.645,
+      "step": 58
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.76e-05,
+      "loss": 0.5641,
+      "step": 87
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6936274509803921,
+      "eval_combined_score": 0.7358012920976561,
+      "eval_f1": 0.7779751332149201,
+      "eval_loss": 0.6025665998458862,
+      "eval_runtime": 0.708,
+      "eval_samples_per_second": 576.282,
+      "eval_steps_per_second": 5.65,
+      "step": 87
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.4682,
+      "step": 116
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7034313725490197,
+      "eval_combined_score": 0.7503756527736722,
+      "eval_f1": 0.7973199329983249,
+      "eval_loss": 0.633918285369873,
+      "eval_runtime": 0.7082,
+      "eval_samples_per_second": 576.107,
+      "eval_steps_per_second": 5.648,
+      "step": 116
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.3677,
+      "step": 145
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6495098039215687,
+      "eval_combined_score": 0.6901033012074886,
+      "eval_f1": 0.7306967984934086,
+      "eval_loss": 0.9408003687858582,
+      "eval_runtime": 0.7111,
+      "eval_samples_per_second": 573.777,
+      "eval_steps_per_second": 5.625,
+      "step": 145
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 3.52e-05,
+      "loss": 0.2183,
+      "step": 174
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6544117647058824,
+      "eval_combined_score": 0.701087814374408,
+      "eval_f1": 0.7477638640429337,
+      "eval_loss": 0.8310534358024597,
+      "eval_runtime": 0.7112,
+      "eval_samples_per_second": 573.68,
+      "eval_steps_per_second": 5.624,
+      "step": 174
+    },
+    {
+      "epoch": 6.0,
+      "step": 174,
+      "total_flos": 3220632565186560.0,
+      "train_loss": 0.48217608736849377,
+      "train_runtime": 133.0395,
+      "train_samples_per_second": 1378.538,
+      "train_steps_per_second": 10.899
+    }
+  ],
+  "max_steps": 1450,
+  "num_train_epochs": 50,
+  "total_flos": 3220632565186560.0,
+  "trial_name": null,
+  "trial_params": null
+}