End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 base_model: gokuls/bert_12_layer_model_v1_complete_training_new_48
 tags:
 - generated_from_trainer
@@ -13,7 +15,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE MNLI
       type: glue
       config: mnli
       split: validation_matched
@@ -21,7 +23,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.31818644931227713
+      value: 0.3522172497965826
 ---
 
 <!-- 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. -->
 
 # hBERTv1_new_pretrain_48_ver2_mnli
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_48](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_48) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_48](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_48) on the GLUE MNLI dataset.
 It achieves the following results on the evaluation set:
 - Loss: 1.0986
-- Accuracy: 0.3182
+- Accuracy: 0.3522
 
 ## Model description
 

all_results.json

@@ -0,0 +1,21 @@
+{
+    "epoch": 9.0,
+    "epoch_mm": 9.0,
+    "eval_accuracy": 0.3544574630667346,
+    "eval_accuracy_mm": 0.3522172497965826,
+    "eval_loss": 1.0985641479492188,
+    "eval_loss_mm": 1.0985745191574097,
+    "eval_runtime": 21.5774,
+    "eval_runtime_mm": 21.611,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 454.875,
+    "eval_samples_per_second_mm": 454.953,
+    "eval_steps_per_second": 7.137,
+    "eval_steps_per_second_mm": 7.126,
+    "train_loss": 1.0989558399371198,
+    "train_runtime": 25003.8725,
+    "train_samples": 392702,
+    "train_samples_per_second": 235.585,
+    "train_steps_per_second": 3.681
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 9.0,
+    "epoch_mm": 9.0,
+    "eval_accuracy": 0.3544574630667346,
+    "eval_accuracy_mm": 0.3522172497965826,
+    "eval_loss": 1.0985641479492188,
+    "eval_loss_mm": 1.0985745191574097,
+    "eval_runtime": 21.5774,
+    "eval_runtime_mm": 21.611,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 454.875,
+    "eval_samples_per_second_mm": 454.953,
+    "eval_steps_per_second": 7.137,
+    "eval_steps_per_second_mm": 7.126
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 9.0,
+    "train_loss": 1.0989558399371198,
+    "train_runtime": 25003.8725,
+    "train_samples": 392702,
+    "train_samples_per_second": 235.585,
+    "train_steps_per_second": 3.681
+}

trainer_state.json

@@ -0,0 +1,163 @@
+{
+  "best_metric": 1.0985641479492188,
+  "best_model_checkpoint": "hBERTv1_new_pretrain_48_ver2_mnli/checkpoint-24544",
+  "epoch": 9.0,
+  "eval_steps": 500,
+  "global_step": 55224,
+  "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": 1.1013,
+      "step": 6136
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.31818644931227713,
+      "eval_loss": 1.0990490913391113,
+      "eval_runtime": 21.8244,
+      "eval_samples_per_second": 449.726,
+      "eval_steps_per_second": 7.056,
+      "step": 6136
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.466666666666667e-05,
+      "loss": 1.0988,
+      "step": 12272
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.31818644931227713,
+      "eval_loss": 1.099389910697937,
+      "eval_runtime": 21.8009,
+      "eval_samples_per_second": 450.21,
+      "eval_steps_per_second": 7.064,
+      "step": 12272
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 1.0987,
+      "step": 18408
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.31818644931227713,
+      "eval_loss": 1.0986260175704956,
+      "eval_runtime": 21.7566,
+      "eval_samples_per_second": 451.128,
+      "eval_steps_per_second": 7.078,
+      "step": 18408
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 2.9333333333333333e-05,
+      "loss": 1.0986,
+      "step": 24544
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": 1.0985641479492188,
+      "eval_runtime": 21.6777,
+      "eval_samples_per_second": 452.769,
+      "eval_steps_per_second": 7.104,
+      "step": 24544
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 2.6666666666666667e-05,
+      "loss": 1.0986,
+      "step": 30680
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": 1.0985902547836304,
+      "eval_runtime": 21.7473,
+      "eval_samples_per_second": 451.321,
+      "eval_steps_per_second": 7.081,
+      "step": 30680
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 2.4e-05,
+      "loss": 1.0986,
+      "step": 36816
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.3273560876209883,
+      "eval_loss": 1.0986045598983765,
+      "eval_runtime": 21.8778,
+      "eval_samples_per_second": 448.627,
+      "eval_steps_per_second": 7.039,
+      "step": 36816
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 2.1333333333333335e-05,
+      "loss": 1.0986,
+      "step": 42952
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": 1.098606824874878,
+      "eval_runtime": 21.6702,
+      "eval_samples_per_second": 452.927,
+      "eval_steps_per_second": 7.107,
+      "step": 42952
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 1.866666666666667e-05,
+      "loss": 1.0986,
+      "step": 49088
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": 1.0986090898513794,
+      "eval_runtime": 21.6888,
+      "eval_samples_per_second": 452.539,
+      "eval_steps_per_second": 7.1,
+      "step": 49088
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 1.6000000000000003e-05,
+      "loss": 1.0986,
+      "step": 55224
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.31818644931227713,
+      "eval_loss": 1.098615050315857,
+      "eval_runtime": 21.6696,
+      "eval_samples_per_second": 452.938,
+      "eval_steps_per_second": 7.107,
+      "step": 55224
+    },
+    {
+      "epoch": 9.0,
+      "step": 55224,
+      "total_flos": 5.2342526347680154e+17,
+      "train_loss": 1.0989558399371198,
+      "train_runtime": 25003.8725,
+      "train_samples_per_second": 235.585,
+      "train_steps_per_second": 3.681
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 92040,
+  "num_train_epochs": 15,
+  "save_steps": 500,
+  "total_flos": 5.2342526347680154e+17,
+  "trial_name": null,
+  "trial_params": null
+}