End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 tags:
 - generated_from_trainer
 datasets:
@@ -12,7 +14,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE MNLI
       type: glue
       config: mnli
       split: validation_matched
@@ -20,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.31818644931227713
+      value: 0.3287225386493084
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -28,10 +30,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hBERTv2_new_pretrain_48_emb_com_mnli
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new_emb_compress_48](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new_emb_compress_48) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new_emb_compress_48](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new_emb_compress_48) on the GLUE MNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0986
-- Accuracy: 0.3182
+- Loss: 1.0979
+- Accuracy: 0.3287
 
 ## Model description
 

all_results.json

@@ -0,0 +1,21 @@
+{
+    "epoch": 10.0,
+    "epoch_mm": 10.0,
+    "eval_accuracy": 0.3329597554763118,
+    "eval_accuracy_mm": 0.3287225386493084,
+    "eval_loss": 1.0977137088775635,
+    "eval_loss_mm": 1.0978821516036987,
+    "eval_runtime": 42.5617,
+    "eval_runtime_mm": 42.8215,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 230.606,
+    "eval_samples_per_second_mm": 229.604,
+    "eval_steps_per_second": 1.809,
+    "eval_steps_per_second_mm": 1.798,
+    "train_loss": 1.0995322733534671,
+    "train_runtime": 52543.2624,
+    "train_samples": 392702,
+    "train_samples_per_second": 373.694,
+    "train_steps_per_second": 2.919
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 10.0,
+    "epoch_mm": 10.0,
+    "eval_accuracy": 0.3329597554763118,
+    "eval_accuracy_mm": 0.3287225386493084,
+    "eval_loss": 1.0977137088775635,
+    "eval_loss_mm": 1.0978821516036987,
+    "eval_runtime": 42.5617,
+    "eval_runtime_mm": 42.8215,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 230.606,
+    "eval_samples_per_second_mm": 229.604,
+    "eval_steps_per_second": 1.809,
+    "eval_steps_per_second_mm": 1.798
+}

logs/events.out.tfevents.1686909795.glendale.42326.17

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:823eb6b1fee113db939d557eb625284a0840e81eebab595bfcafcc9dde843a33
+size 698

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 10.0,
+    "train_loss": 1.0995322733534671,
+    "train_runtime": 52543.2624,
+    "train_samples": 392702,
+    "train_samples_per_second": 373.694,
+    "train_steps_per_second": 2.919
+}

trainer_state.json

@@ -0,0 +1,175 @@
+{
+  "best_metric": 1.0977137088775635,
+  "best_model_checkpoint": "hBERTv2_new_pretrain_48_emb_com_mnli/checkpoint-15340",
+  "epoch": 10.0,
+  "global_step": 30680,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9200000000000004e-05,
+      "loss": 1.1059,
+      "step": 3068
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.31818644931227713,
+      "eval_loss": 1.103637456893921,
+      "eval_runtime": 43.1654,
+      "eval_samples_per_second": 227.381,
+      "eval_steps_per_second": 1.784,
+      "step": 3068
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 1.1014,
+      "step": 6136
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.3273560876209883,
+      "eval_loss": 1.1026175022125244,
+      "eval_runtime": 43.1828,
+      "eval_samples_per_second": 227.289,
+      "eval_steps_per_second": 1.783,
+      "step": 6136
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.76e-05,
+      "loss": 1.0997,
+      "step": 9204
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.31818644931227713,
+      "eval_loss": 1.1008087396621704,
+      "eval_runtime": 43.0883,
+      "eval_samples_per_second": 227.788,
+      "eval_steps_per_second": 1.787,
+      "step": 9204
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 1.0982,
+      "step": 12272
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.33112582781456956,
+      "eval_loss": 1.0977925062179565,
+      "eval_runtime": 43.0333,
+      "eval_samples_per_second": 228.079,
+      "eval_steps_per_second": 1.789,
+      "step": 12272
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.6e-05,
+      "loss": 1.0978,
+      "step": 15340
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.3329597554763118,
+      "eval_loss": 1.0977137088775635,
+      "eval_runtime": 43.058,
+      "eval_samples_per_second": 227.948,
+      "eval_steps_per_second": 1.788,
+      "step": 15340
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 3.52e-05,
+      "loss": 1.0976,
+      "step": 18408
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": 1.0984772443771362,
+      "eval_runtime": 43.1911,
+      "eval_samples_per_second": 227.246,
+      "eval_steps_per_second": 1.783,
+      "step": 18408
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 3.44e-05,
+      "loss": 1.0987,
+      "step": 21476
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.3182883341823739,
+      "eval_loss": 1.0986559391021729,
+      "eval_runtime": 42.9975,
+      "eval_samples_per_second": 228.269,
+      "eval_steps_per_second": 1.791,
+      "step": 21476
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 1.0987,
+      "step": 24544
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": 1.098609447479248,
+      "eval_runtime": 43.0102,
+      "eval_samples_per_second": 228.202,
+      "eval_steps_per_second": 1.79,
+      "step": 24544
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 3.28e-05,
+      "loss": 1.0986,
+      "step": 27612
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": 1.0986100435256958,
+      "eval_runtime": 43.1085,
+      "eval_samples_per_second": 227.682,
+      "eval_steps_per_second": 1.786,
+      "step": 27612
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 1.0987,
+      "step": 30680
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.31818644931227713,
+      "eval_loss": 1.0986064672470093,
+      "eval_runtime": 43.0847,
+      "eval_samples_per_second": 227.807,
+      "eval_steps_per_second": 1.787,
+      "step": 30680
+    },
+    {
+      "epoch": 10.0,
+      "step": 30680,
+      "total_flos": 5.752713556000768e+17,
+      "train_loss": 1.0995322733534671,
+      "train_runtime": 52543.2624,
+      "train_samples_per_second": 373.694,
+      "train_steps_per_second": 2.919
+    }
+  ],
+  "max_steps": 153400,
+  "num_train_epochs": 50,
+  "total_flos": 5.752713556000768e+17,
+  "trial_name": null,
+  "trial_params": null
+}