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 WNLI
       type: glue
       config: wnli
       split: validation
@@ -21,7 +23,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.5633802816901409
+      value: 0.43661971830985913
 ---
 
 <!-- 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_wnli
 
-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 WNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7029
-- Accuracy: 0.5634
+- Loss: 0.7002
+- Accuracy: 0.4366
 
 ## Model description
 

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.43661971830985913,
+    "eval_loss": 0.7001590728759766,
+    "eval_runtime": 0.1637,
+    "eval_samples": 71,
+    "eval_samples_per_second": 433.651,
+    "eval_steps_per_second": 12.216,
+    "train_loss": 0.7120633738381522,
+    "train_runtime": 51.357,
+    "train_samples": 635,
+    "train_samples_per_second": 185.466,
+    "train_steps_per_second": 2.921
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.43661971830985913,
+    "eval_loss": 0.7001590728759766,
+    "eval_runtime": 0.1637,
+    "eval_samples": 71,
+    "eval_samples_per_second": 433.651,
+    "eval_steps_per_second": 12.216
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 7.0,
+    "train_loss": 0.7120633738381522,
+    "train_runtime": 51.357,
+    "train_samples": 635,
+    "train_samples_per_second": 185.466,
+    "train_steps_per_second": 2.921
+}

trainer_state.json

@@ -0,0 +1,133 @@
+{
+  "best_metric": 0.7001590728759766,
+  "best_model_checkpoint": "hBERTv1_new_pretrain_48_ver2_wnli/checkpoint-20",
+  "epoch": 7.0,
+  "eval_steps": 500,
+  "global_step": 70,
+  "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": 0.7332,
+      "step": 10
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 0.7386435270309448,
+      "eval_runtime": 0.1611,
+      "eval_samples_per_second": 440.599,
+      "eval_steps_per_second": 12.411,
+      "step": 10
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.466666666666667e-05,
+      "loss": 0.7093,
+      "step": 20
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 0.7001590728759766,
+      "eval_runtime": 0.1609,
+      "eval_samples_per_second": 441.266,
+      "eval_steps_per_second": 12.43,
+      "step": 20
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.7175,
+      "step": 30
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 0.7295132875442505,
+      "eval_runtime": 0.1605,
+      "eval_samples_per_second": 442.375,
+      "eval_steps_per_second": 12.461,
+      "step": 30
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 2.9333333333333333e-05,
+      "loss": 0.7044,
+      "step": 40
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 0.7007160186767578,
+      "eval_runtime": 0.1614,
+      "eval_samples_per_second": 440.026,
+      "eval_steps_per_second": 12.395,
+      "step": 40
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 2.6666666666666667e-05,
+      "loss": 0.6906,
+      "step": 50
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 0.7484099864959717,
+      "eval_runtime": 0.1609,
+      "eval_samples_per_second": 441.232,
+      "eval_steps_per_second": 12.429,
+      "step": 50
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 2.4e-05,
+      "loss": 0.7095,
+      "step": 60
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 0.7176557183265686,
+      "eval_runtime": 0.1622,
+      "eval_samples_per_second": 437.711,
+      "eval_steps_per_second": 12.33,
+      "step": 60
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 2.1333333333333335e-05,
+      "loss": 0.7201,
+      "step": 70
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.7029308676719666,
+      "eval_runtime": 0.1607,
+      "eval_samples_per_second": 441.951,
+      "eval_steps_per_second": 12.449,
+      "step": 70
+    },
+    {
+      "epoch": 7.0,
+      "step": 70,
+      "total_flos": 658290182717440.0,
+      "train_loss": 0.7120633738381522,
+      "train_runtime": 51.357,
+      "train_samples_per_second": 185.466,
+      "train_steps_per_second": 2.921
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 150,
+  "num_train_epochs": 15,
+  "save_steps": 500,
+  "total_flos": 658290182717440.0,
+  "trial_name": null,
+  "trial_params": null
+}