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: 0.3161764705882353
+      value: 0.6838235294117647
     - name: F1
       type: f1
-      value: 0.0
+      value: 0.8122270742358079
 ---
 
 <!-- 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_mrpc
 
-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 MRPC dataset.
 It achieves the following results on the evaluation set:
-- Loss: 14.0184
-- Accuracy: 0.3162
-- F1: 0.0
-- Combined Score: 0.1581
+- Loss: 1.8707
+- Accuracy: 0.6838
+- F1: 0.8122
+- Combined Score: 0.7480
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.6838235294117647,
+    "eval_combined_score": 0.7480253018237863,
+    "eval_f1": 0.8122270742358079,
+    "eval_loss": 1.870712399482727,
+    "eval_runtime": 0.7206,
+    "eval_samples": 408,
+    "eval_samples_per_second": 566.174,
+    "eval_steps_per_second": 5.551,
+    "train_loss": 9.304771160257273,
+    "train_runtime": 158.2814,
+    "train_samples": 3668,
+    "train_samples_per_second": 1158.696,
+    "train_steps_per_second": 9.161
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.6838235294117647,
+    "eval_combined_score": 0.7480253018237863,
+    "eval_f1": 0.8122270742358079,
+    "eval_loss": 1.870712399482727,
+    "eval_runtime": 0.7206,
+    "eval_samples": 408,
+    "eval_samples_per_second": 566.174,
+    "eval_steps_per_second": 5.551
+}

logs/events.out.tfevents.1685526578.serv-3318.2595329.8

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:29b976d748951080f20b51b468cb0d6237514f44d1047f9f87b53b1f35e5881d
+size 467

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 7.0,
+    "train_loss": 9.304771160257273,
+    "train_runtime": 158.2814,
+    "train_samples": 3668,
+    "train_samples_per_second": 1158.696,
+    "train_steps_per_second": 9.161
+}

trainer_state.json

@@ -0,0 +1,144 @@
+{
+  "best_metric": 1.870712399482727,
+  "best_model_checkpoint": "hBERTv2_new_pretrain_mrpc/checkpoint-58",
+  "epoch": 7.0,
+  "global_step": 203,
+  "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": 8.6576,
+      "step": 29
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 3.519158124923706,
+      "eval_runtime": 0.7223,
+      "eval_samples_per_second": 564.843,
+      "eval_steps_per_second": 5.538,
+      "step": 29
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 0.00048,
+      "loss": 9.1979,
+      "step": 58
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 1.870712399482727,
+      "eval_runtime": 0.7217,
+      "eval_samples_per_second": 565.309,
+      "eval_steps_per_second": 5.542,
+      "step": 58
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 0.00047,
+      "loss": 8.8322,
+      "step": 87
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 3.970895767211914,
+      "eval_runtime": 0.7174,
+      "eval_samples_per_second": 568.684,
+      "eval_steps_per_second": 5.575,
+      "step": 87
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 0.00046,
+      "loss": 8.8403,
+      "step": 116
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 3.3888018131256104,
+      "eval_runtime": 0.7217,
+      "eval_samples_per_second": 565.312,
+      "eval_steps_per_second": 5.542,
+      "step": 116
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 0.00045000000000000004,
+      "loss": 8.3428,
+      "step": 145
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 2.3995490074157715,
+      "eval_runtime": 0.7172,
+      "eval_samples_per_second": 568.866,
+      "eval_steps_per_second": 5.577,
+      "step": 145
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 0.00044,
+      "loss": 8.565,
+      "step": 174
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 5.375,
+      "eval_runtime": 0.718,
+      "eval_samples_per_second": 568.237,
+      "eval_steps_per_second": 5.571,
+      "step": 174
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 0.00043,
+      "loss": 12.6976,
+      "step": 203
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.3161764705882353,
+      "eval_combined_score": 0.15808823529411764,
+      "eval_f1": 0.0,
+      "eval_loss": 14.01838207244873,
+      "eval_runtime": 0.7194,
+      "eval_samples_per_second": 567.161,
+      "eval_steps_per_second": 5.56,
+      "step": 203
+    },
+    {
+      "epoch": 7.0,
+      "step": 203,
+      "total_flos": 3757404659384320.0,
+      "train_loss": 9.304771160257273,
+      "train_runtime": 158.2814,
+      "train_samples_per_second": 1158.696,
+      "train_steps_per_second": 9.161
+    }
+  ],
+  "max_steps": 1450,
+  "num_train_epochs": 50,
+  "total_flos": 3757404659384320.0,
+  "trial_name": null,
+  "trial_params": null
+}