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: 0.6838235294117647
+      value: 0.6862745098039216
     - name: F1
       type: f1
-      value: 0.7809847198641766
+      value: 0.7999999999999999
 ---
 
 <!-- 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. -->
 
 # hBERTv1_mrpc
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1](https://huggingface.co/gokuls/bert_12_layer_model_v1) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1](https://huggingface.co/gokuls/bert_12_layer_model_v1) on the GLUE MRPC dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.9708
-- Accuracy: 0.6838
-- F1: 0.7810
-- Combined Score: 0.7324
+- Loss: 0.6051
+- Accuracy: 0.6863
+- F1: 0.8000
+- Combined Score: 0.7431
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 12.0,
+    "eval_accuracy": 0.6862745098039216,
+    "eval_combined_score": 0.7431372549019608,
+    "eval_f1": 0.7999999999999999,
+    "eval_loss": 0.6050820350646973,
+    "eval_runtime": 0.5403,
+    "eval_samples": 408,
+    "eval_samples_per_second": 755.096,
+    "eval_steps_per_second": 3.701,
+    "train_loss": 0.4964691546228197,
+    "train_runtime": 203.7797,
+    "train_samples": 3668,
+    "train_samples_per_second": 899.991,
+    "train_steps_per_second": 3.68
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 12.0,
+    "eval_accuracy": 0.6862745098039216,
+    "eval_combined_score": 0.7431372549019608,
+    "eval_f1": 0.7999999999999999,
+    "eval_loss": 0.6050820350646973,
+    "eval_runtime": 0.5403,
+    "eval_samples": 408,
+    "eval_samples_per_second": 755.096,
+    "eval_steps_per_second": 3.701
+}

logs/events.out.tfevents.1677681169.serv-9220.1178704.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f2d7267f8afbe59b4250732729bd83828a8faad617fbd82aed01440282af639
+size 467

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 12.0,
+    "train_loss": 0.4964691546228197,
+    "train_runtime": 203.7797,
+    "train_samples": 3668,
+    "train_samples_per_second": 899.991,
+    "train_steps_per_second": 3.68
+}

trainer_state.json

@@ -0,0 +1,229 @@
+{
+  "best_metric": 0.6050820350646973,
+  "best_model_checkpoint": "hBERTv1_mrpc/checkpoint-105",
+  "epoch": 12.0,
+  "global_step": 180,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.9e-05,
+      "loss": 0.6536,
+      "step": 15
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6242610216140747,
+      "eval_runtime": 0.5132,
+      "eval_samples_per_second": 794.982,
+      "eval_steps_per_second": 3.897,
+      "step": 15
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.8e-05,
+      "loss": 0.6275,
+      "step": 30
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.7009803921568627,
+      "eval_combined_score": 0.7563543936092956,
+      "eval_f1": 0.8117283950617284,
+      "eval_loss": 0.6174396872520447,
+      "eval_runtime": 0.511,
+      "eval_samples_per_second": 798.461,
+      "eval_steps_per_second": 3.914,
+      "step": 30
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7e-05,
+      "loss": 0.6129,
+      "step": 45
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.696078431372549,
+      "eval_combined_score": 0.7571301247771836,
+      "eval_f1": 0.8181818181818181,
+      "eval_loss": 0.6089469194412231,
+      "eval_runtime": 0.5125,
+      "eval_samples_per_second": 796.115,
+      "eval_steps_per_second": 3.903,
+      "step": 45
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.6087,
+      "step": 60
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6887254901960784,
+      "eval_combined_score": 0.7508428629183632,
+      "eval_f1": 0.8129602356406479,
+      "eval_loss": 0.606163740158081,
+      "eval_runtime": 0.5136,
+      "eval_samples_per_second": 794.368,
+      "eval_steps_per_second": 3.894,
+      "step": 60
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.5e-05,
+      "loss": 0.5939,
+      "step": 75
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6862745098039216,
+      "eval_combined_score": 0.739911448450348,
+      "eval_f1": 0.7935483870967743,
+      "eval_loss": 0.6104128956794739,
+      "eval_runtime": 0.5141,
+      "eval_samples_per_second": 793.575,
+      "eval_steps_per_second": 3.89,
+      "step": 75
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.5707,
+      "step": 90
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.7083333333333334,
+      "eval_combined_score": 0.7633269720101782,
+      "eval_f1": 0.818320610687023,
+      "eval_loss": 0.6184039115905762,
+      "eval_runtime": 0.5141,
+      "eval_samples_per_second": 793.682,
+      "eval_steps_per_second": 3.891,
+      "step": 90
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.3e-05,
+      "loss": 0.5426,
+      "step": 105
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6862745098039216,
+      "eval_combined_score": 0.7431372549019608,
+      "eval_f1": 0.7999999999999999,
+      "eval_loss": 0.6050820350646973,
+      "eval_runtime": 0.5141,
+      "eval_samples_per_second": 793.592,
+      "eval_steps_per_second": 3.89,
+      "step": 105
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.2e-05,
+      "loss": 0.4819,
+      "step": 120
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6936274509803921,
+      "eval_combined_score": 0.7477645971225257,
+      "eval_f1": 0.8019017432646594,
+      "eval_loss": 0.6560007929801941,
+      "eval_runtime": 0.515,
+      "eval_samples_per_second": 792.192,
+      "eval_steps_per_second": 3.883,
+      "step": 120
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 4.1e-05,
+      "loss": 0.4279,
+      "step": 135
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.6887254901960784,
+      "eval_combined_score": 0.7282749937269242,
+      "eval_f1": 0.7678244972577698,
+      "eval_loss": 0.6672889590263367,
+      "eval_runtime": 0.5142,
+      "eval_samples_per_second": 793.515,
+      "eval_steps_per_second": 3.89,
+      "step": 135
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 4e-05,
+      "loss": 0.3374,
+      "step": 150
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.6862745098039216,
+      "eval_combined_score": 0.7382192221150756,
+      "eval_f1": 0.7901639344262296,
+      "eval_loss": 0.8091686367988586,
+      "eval_runtime": 0.5155,
+      "eval_samples_per_second": 791.402,
+      "eval_steps_per_second": 3.879,
+      "step": 150
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 3.9000000000000006e-05,
+      "loss": 0.2789,
+      "step": 165
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.6887254901960784,
+      "eval_combined_score": 0.741110712577714,
+      "eval_f1": 0.7934959349593496,
+      "eval_loss": 0.9342193603515625,
+      "eval_runtime": 0.5139,
+      "eval_samples_per_second": 793.922,
+      "eval_steps_per_second": 3.892,
+      "step": 165
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 3.8e-05,
+      "loss": 0.2216,
+      "step": 180
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7324041246379707,
+      "eval_f1": 0.7809847198641766,
+      "eval_loss": 0.9707943201065063,
+      "eval_runtime": 0.5142,
+      "eval_samples_per_second": 793.529,
+      "eval_steps_per_second": 3.89,
+      "step": 180
+    },
+    {
+      "epoch": 12.0,
+      "step": 180,
+      "total_flos": 5559706315653120.0,
+      "train_loss": 0.4964691546228197,
+      "train_runtime": 203.7797,
+      "train_samples_per_second": 899.991,
+      "train_steps_per_second": 3.68
+    }
+  ],
+  "max_steps": 750,
+  "num_train_epochs": 50,
+  "total_flos": 5559706315653120.0,
+  "trial_name": null,
+  "trial_params": null
+}