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: 0.678921568627451
+      value: 0.696078431372549
     - name: F1
       type: f1
-      value: 0.7713787085514833
+      value: 0.7912457912457912
 ---
 
 <!-- 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_48_mrpc
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new_wt_init_48](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new_wt_init_48) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new_wt_init_48](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new_wt_init_48) on the GLUE MRPC dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.1768
-- Accuracy: 0.6789
-- F1: 0.7714
-- Combined Score: 0.7252
+- Loss: 0.5938
+- Accuracy: 0.6961
+- F1: 0.7912
+- Combined Score: 0.7437
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.696078431372549,
+    "eval_combined_score": 0.7436621113091701,
+    "eval_f1": 0.7912457912457912,
+    "eval_loss": 0.5938104391098022,
+    "eval_runtime": 0.7475,
+    "eval_samples": 408,
+    "eval_samples_per_second": 545.847,
+    "eval_steps_per_second": 5.351,
+    "train_loss": 0.438173795568532,
+    "train_runtime": 185.9399,
+    "train_samples": 3668,
+    "train_samples_per_second": 986.34,
+    "train_steps_per_second": 7.798
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.696078431372549,
+    "eval_combined_score": 0.7436621113091701,
+    "eval_f1": 0.7912457912457912,
+    "eval_loss": 0.5938104391098022,
+    "eval_runtime": 0.7475,
+    "eval_samples": 408,
+    "eval_samples_per_second": 545.847,
+    "eval_steps_per_second": 5.351
+}

logs/events.out.tfevents.1686034728.serv-9207.2390652.8

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a0d12fb2b0449197ab1fbee0999dbe855bea97989101cff4e704980e27dc7d92
+size 467

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 8.0,
+    "train_loss": 0.438173795568532,
+    "train_runtime": 185.9399,
+    "train_samples": 3668,
+    "train_samples_per_second": 986.34,
+    "train_steps_per_second": 7.798
+}

trainer_state.json

@@ -0,0 +1,161 @@
+{
+  "best_metric": 0.5938104391098022,
+  "best_model_checkpoint": "hBERTv2_new_pretrain_w_init_48_mrpc/checkpoint-87",
+  "epoch": 8.0,
+  "global_step": 232,
+  "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": 0.6532,
+      "step": 29
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7454992086520706,
+      "eval_f1": 0.8071748878923767,
+      "eval_loss": 0.6271159052848816,
+      "eval_runtime": 0.7188,
+      "eval_samples_per_second": 567.628,
+      "eval_steps_per_second": 5.565,
+      "step": 29
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.6168,
+      "step": 58
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6813725490196079,
+      "eval_combined_score": 0.7436713491366695,
+      "eval_f1": 0.8059701492537312,
+      "eval_loss": 0.5971189141273499,
+      "eval_runtime": 0.7238,
+      "eval_samples_per_second": 563.691,
+      "eval_steps_per_second": 5.526,
+      "step": 58
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.76e-05,
+      "loss": 0.5766,
+      "step": 87
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.696078431372549,
+      "eval_combined_score": 0.7436621113091701,
+      "eval_f1": 0.7912457912457912,
+      "eval_loss": 0.5938104391098022,
+      "eval_runtime": 0.721,
+      "eval_samples_per_second": 565.914,
+      "eval_steps_per_second": 5.548,
+      "step": 87
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.5304,
+      "step": 116
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7058823529411765,
+      "eval_combined_score": 0.7549019607843137,
+      "eval_f1": 0.803921568627451,
+      "eval_loss": 0.6173788905143738,
+      "eval_runtime": 0.7253,
+      "eval_samples_per_second": 562.533,
+      "eval_steps_per_second": 5.515,
+      "step": 116
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.4622,
+      "step": 145
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.678921568627451,
+      "eval_combined_score": 0.7271108700770188,
+      "eval_f1": 0.7753001715265866,
+      "eval_loss": 0.6872721910476685,
+      "eval_runtime": 0.7246,
+      "eval_samples_per_second": 563.072,
+      "eval_steps_per_second": 5.52,
+      "step": 145
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 3.52e-05,
+      "loss": 0.3228,
+      "step": 174
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6887254901960784,
+      "eval_combined_score": 0.7390559457613892,
+      "eval_f1": 0.7893864013266999,
+      "eval_loss": 0.7266761064529419,
+      "eval_runtime": 0.7297,
+      "eval_samples_per_second": 559.161,
+      "eval_steps_per_second": 5.482,
+      "step": 174
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 3.44e-05,
+      "loss": 0.2028,
+      "step": 203
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.7009803921568627,
+      "eval_combined_score": 0.744218419074947,
+      "eval_f1": 0.7874564459930313,
+      "eval_loss": 0.9771444201469421,
+      "eval_runtime": 0.7295,
+      "eval_samples_per_second": 559.292,
+      "eval_steps_per_second": 5.483,
+      "step": 203
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 0.1406,
+      "step": 232
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.678921568627451,
+      "eval_combined_score": 0.7251501385894672,
+      "eval_f1": 0.7713787085514833,
+      "eval_loss": 1.1767858266830444,
+      "eval_runtime": 0.7394,
+      "eval_samples_per_second": 551.763,
+      "eval_steps_per_second": 5.409,
+      "step": 232
+    },
+    {
+      "epoch": 8.0,
+      "step": 232,
+      "total_flos": 4294176753582080.0,
+      "train_loss": 0.438173795568532,
+      "train_runtime": 185.9399,
+      "train_samples_per_second": 986.34,
+      "train_steps_per_second": 7.798
+    }
+  ],
+  "max_steps": 1450,
+  "num_train_epochs": 50,
+  "total_flos": 4294176753582080.0,
+  "trial_name": null,
+  "trial_params": null
+}