End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 base_model: gokuls/bert_12_layer_model_v2_complete_training_new_wt_init_48
 tags:
 - generated_from_trainer
@@ -13,7 +15,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE QNLI
       type: glue
       config: qnli
       split: validation
@@ -21,7 +23,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.4946000366099213
+      value: 0.5053999633900788
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -29,10 +31,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hBERTv2_new_pretrain_w_init_48_ver2_qnli
 
-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 QNLI dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.6931
-- Accuracy: 0.4946
+- Accuracy: 0.5054
 
 ## Model description
 

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 13.0,
+    "eval_accuracy": 0.5053999633900788,
+    "eval_loss": 0.6930909752845764,
+    "eval_runtime": 11.3001,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 483.446,
+    "eval_steps_per_second": 7.611,
+    "train_loss": 0.6940912691522073,
+    "train_runtime": 9216.7254,
+    "train_samples": 104743,
+    "train_samples_per_second": 170.467,
+    "train_steps_per_second": 2.664
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 13.0,
+    "eval_accuracy": 0.5053999633900788,
+    "eval_loss": 0.6930909752845764,
+    "eval_runtime": 11.3001,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 483.446,
+    "eval_steps_per_second": 7.611
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 13.0,
+    "train_loss": 0.6940912691522073,
+    "train_runtime": 9216.7254,
+    "train_samples": 104743,
+    "train_samples_per_second": 170.467,
+    "train_steps_per_second": 2.664
+}

trainer_state.json

@@ -0,0 +1,223 @@
+{
+  "best_metric": 0.6930909752845764,
+  "best_model_checkpoint": "hBERTv2_new_pretrain_w_init_48_ver2_qnli/checkpoint-13096",
+  "epoch": 13.0,
+  "eval_steps": 500,
+  "global_step": 21281,
+  "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": 0.7008,
+      "step": 1637
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 0.6943343281745911,
+      "eval_runtime": 11.5343,
+      "eval_samples_per_second": 473.629,
+      "eval_steps_per_second": 7.456,
+      "step": 1637
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.466666666666667e-05,
+      "loss": 0.6946,
+      "step": 3274
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 0.6931345462799072,
+      "eval_runtime": 11.5302,
+      "eval_samples_per_second": 473.798,
+      "eval_steps_per_second": 7.459,
+      "step": 3274
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.6938,
+      "step": 4911
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.4946000366099213,
+      "eval_loss": 0.6931858062744141,
+      "eval_runtime": 11.5515,
+      "eval_samples_per_second": 472.924,
+      "eval_steps_per_second": 7.445,
+      "step": 4911
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 2.9333333333333333e-05,
+      "loss": 0.6943,
+      "step": 6548
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 0.6933774352073669,
+      "eval_runtime": 11.5393,
+      "eval_samples_per_second": 473.424,
+      "eval_steps_per_second": 7.453,
+      "step": 6548
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 2.6666666666666667e-05,
+      "loss": 0.694,
+      "step": 8185
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.4946000366099213,
+      "eval_loss": 0.6933080554008484,
+      "eval_runtime": 11.5044,
+      "eval_samples_per_second": 474.86,
+      "eval_steps_per_second": 7.475,
+      "step": 8185
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 2.4e-05,
+      "loss": 0.6932,
+      "step": 9822
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 0.6930915713310242,
+      "eval_runtime": 11.4731,
+      "eval_samples_per_second": 476.158,
+      "eval_steps_per_second": 7.496,
+      "step": 9822
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 2.1333333333333335e-05,
+      "loss": 0.6934,
+      "step": 11459
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 0.6931099891662598,
+      "eval_runtime": 11.5126,
+      "eval_samples_per_second": 474.522,
+      "eval_steps_per_second": 7.47,
+      "step": 11459
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 1.866666666666667e-05,
+      "loss": 0.6932,
+      "step": 13096
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 0.6930909752845764,
+      "eval_runtime": 11.496,
+      "eval_samples_per_second": 475.207,
+      "eval_steps_per_second": 7.481,
+      "step": 13096
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 1.6000000000000003e-05,
+      "loss": 0.6932,
+      "step": 14733
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.4946000366099213,
+      "eval_loss": 0.6931536197662354,
+      "eval_runtime": 11.4976,
+      "eval_samples_per_second": 475.144,
+      "eval_steps_per_second": 7.48,
+      "step": 14733
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 1.3333333333333333e-05,
+      "loss": 0.6932,
+      "step": 16370
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.4946000366099213,
+      "eval_loss": 0.6932971477508545,
+      "eval_runtime": 11.5141,
+      "eval_samples_per_second": 474.463,
+      "eval_steps_per_second": 7.469,
+      "step": 16370
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 1.0666666666666667e-05,
+      "loss": 0.6932,
+      "step": 18007
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 0.6931360363960266,
+      "eval_runtime": 11.5044,
+      "eval_samples_per_second": 474.863,
+      "eval_steps_per_second": 7.475,
+      "step": 18007
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 0.6932,
+      "step": 19644
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 0.6931059956550598,
+      "eval_runtime": 11.4771,
+      "eval_samples_per_second": 475.991,
+      "eval_steps_per_second": 7.493,
+      "step": 19644
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 5.333333333333334e-06,
+      "loss": 0.6932,
+      "step": 21281
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.4946000366099213,
+      "eval_loss": 0.6931484937667847,
+      "eval_runtime": 11.5033,
+      "eval_samples_per_second": 474.906,
+      "eval_steps_per_second": 7.476,
+      "step": 21281
+    },
+    {
+      "epoch": 13.0,
+      "step": 21281,
+      "total_flos": 1.99264049920213e+17,
+      "train_loss": 0.6940912691522073,
+      "train_runtime": 9216.7254,
+      "train_samples_per_second": 170.467,
+      "train_steps_per_second": 2.664
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 24555,
+  "num_train_epochs": 15,
+  "save_steps": 500,
+  "total_flos": 1.99264049920213e+17,
+  "trial_name": null,
+  "trial_params": null
+}