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 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. -->
 
 # hBERTv2_new_pretrain_w_init_48_ver2_mnli
 
-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 MNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0986
-- Accuracy: 0.3182
+- Loss: 1.0985
+- 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": 1.0985355377197266,
+    "eval_loss_mm": 1.0985404253005981,
+    "eval_runtime": 20.4507,
+    "eval_runtime_mm": 20.4459,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 479.935,
+    "eval_samples_per_second_mm": 480.878,
+    "eval_steps_per_second": 7.53,
+    "eval_steps_per_second_mm": 7.532,
+    "train_loss": 1.0990169016188613,
+    "train_runtime": 23433.9477,
+    "train_samples": 392702,
+    "train_samples_per_second": 251.367,
+    "train_steps_per_second": 3.928
+}

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": 1.0985355377197266,
+    "eval_loss_mm": 1.0985404253005981,
+    "eval_runtime": 20.4507,
+    "eval_runtime_mm": 20.4459,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 479.935,
+    "eval_samples_per_second_mm": 480.878,
+    "eval_steps_per_second": 7.53,
+    "eval_steps_per_second_mm": 7.532
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 9.0,
+    "train_loss": 1.0990169016188613,
+    "train_runtime": 23433.9477,
+    "train_samples": 392702,
+    "train_samples_per_second": 251.367,
+    "train_steps_per_second": 3.928
+}

trainer_state.json

@@ -0,0 +1,163 @@
+{
+  "best_metric": 1.0985355377197266,
+  "best_model_checkpoint": "hBERTv2_new_pretrain_w_init_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": 1.1017,
+      "step": 6136
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.31818644931227713,
+      "eval_loss": 1.099240779876709,
+      "eval_runtime": 20.6002,
+      "eval_samples_per_second": 476.452,
+      "eval_steps_per_second": 7.476,
+      "step": 6136
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.466666666666667e-05,
+      "loss": 1.0988,
+      "step": 12272
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.31818644931227713,
+      "eval_loss": 1.0993497371673584,
+      "eval_runtime": 20.656,
+      "eval_samples_per_second": 475.165,
+      "eval_steps_per_second": 7.455,
+      "step": 12272
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 1.0987,
+      "step": 18408
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.31818644931227713,
+      "eval_loss": 1.0986136198043823,
+      "eval_runtime": 20.721,
+      "eval_samples_per_second": 473.674,
+      "eval_steps_per_second": 7.432,
+      "step": 18408
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 2.9333333333333333e-05,
+      "loss": 1.0987,
+      "step": 24544
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": 1.0985355377197266,
+      "eval_runtime": 20.7461,
+      "eval_samples_per_second": 473.1,
+      "eval_steps_per_second": 7.423,
+      "step": 24544
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 2.6666666666666667e-05,
+      "loss": 1.0987,
+      "step": 30680
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.31818644931227713,
+      "eval_loss": 1.098646640777588,
+      "eval_runtime": 20.7299,
+      "eval_samples_per_second": 473.47,
+      "eval_steps_per_second": 7.429,
+      "step": 30680
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 2.4e-05,
+      "loss": 1.0986,
+      "step": 36816
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.3273560876209883,
+      "eval_loss": 1.098628282546997,
+      "eval_runtime": 20.7789,
+      "eval_samples_per_second": 472.354,
+      "eval_steps_per_second": 7.411,
+      "step": 36816
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 2.1333333333333335e-05,
+      "loss": 1.0986,
+      "step": 42952
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": 1.098601222038269,
+      "eval_runtime": 20.7327,
+      "eval_samples_per_second": 473.407,
+      "eval_steps_per_second": 7.428,
+      "step": 42952
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 1.866666666666667e-05,
+      "loss": 1.0986,
+      "step": 49088
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": 1.0986049175262451,
+      "eval_runtime": 20.6966,
+      "eval_samples_per_second": 474.232,
+      "eval_steps_per_second": 7.441,
+      "step": 49088
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 1.6000000000000003e-05,
+      "loss": 1.0986,
+      "step": 55224
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.31818644931227713,
+      "eval_loss": 1.098617434501648,
+      "eval_runtime": 20.702,
+      "eval_samples_per_second": 474.11,
+      "eval_steps_per_second": 7.439,
+      "step": 55224
+    },
+    {
+      "epoch": 9.0,
+      "step": 55224,
+      "total_flos": 5.1721333631247974e+17,
+      "train_loss": 1.0990169016188613,
+      "train_runtime": 23433.9477,
+      "train_samples_per_second": 251.367,
+      "train_steps_per_second": 3.928
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 92040,
+  "num_train_epochs": 15,
+  "save_steps": 500,
+  "total_flos": 5.1721333631247974e+17,
+  "trial_name": null,
+  "trial_params": null
+}