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 RTE
       type: glue
       config: rte
       split: validation
@@ -20,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.4729241877256318
+      value: 0.5270758122743683
 ---
 
 <!-- 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. -->
 
 # hBERTv1_new_pretrain_48_KD_rte
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_48_KD](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_48_KD) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_48_KD](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_48_KD) on the GLUE RTE dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7112
-- Accuracy: 0.4729
+- Loss: 0.6917
+- Accuracy: 0.5271
 
 ## Model description
 

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 6.0,
+    "eval_accuracy": 0.5270758122743683,
+    "eval_loss": 0.691717803478241,
+    "eval_runtime": 0.642,
+    "eval_samples": 277,
+    "eval_samples_per_second": 431.471,
+    "eval_steps_per_second": 4.673,
+    "train_loss": 0.7045625448226929,
+    "train_runtime": 133.6091,
+    "train_samples": 2490,
+    "train_samples_per_second": 931.822,
+    "train_steps_per_second": 7.485
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 6.0,
+    "eval_accuracy": 0.5270758122743683,
+    "eval_loss": 0.691717803478241,
+    "eval_runtime": 0.642,
+    "eval_samples": 277,
+    "eval_samples_per_second": 431.471,
+    "eval_steps_per_second": 4.673
+}

logs/events.out.tfevents.1686791944.serv-9207.3372160.11

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e99ddc23542fe34399771e37ed4453decd6532502d6365247f946fbe8b8883aa
+size 357

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 6.0,
+    "train_loss": 0.7045625448226929,
+    "train_runtime": 133.6091,
+    "train_samples": 2490,
+    "train_samples_per_second": 931.822,
+    "train_steps_per_second": 7.485
+}

trainer_state.json

@@ -0,0 +1,115 @@
+{
+  "best_metric": 0.691717803478241,
+  "best_model_checkpoint": "hBERTv1_new_pretrain_48_KD_rte/checkpoint-20",
+  "epoch": 6.0,
+  "global_step": 120,
+  "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": 0.7295,
+      "step": 20
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5270758122743683,
+      "eval_loss": 0.691717803478241,
+      "eval_runtime": 0.5953,
+      "eval_samples_per_second": 465.339,
+      "eval_steps_per_second": 5.04,
+      "step": 20
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.7039,
+      "step": 40
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.4729241877256318,
+      "eval_loss": 0.6931992769241333,
+      "eval_runtime": 0.5885,
+      "eval_samples_per_second": 470.721,
+      "eval_steps_per_second": 5.098,
+      "step": 40
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.76e-05,
+      "loss": 0.6998,
+      "step": 60
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.4729241877256318,
+      "eval_loss": 0.6956332921981812,
+      "eval_runtime": 0.5901,
+      "eval_samples_per_second": 469.427,
+      "eval_steps_per_second": 5.084,
+      "step": 60
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.6954,
+      "step": 80
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.4729241877256318,
+      "eval_loss": 0.6959973573684692,
+      "eval_runtime": 0.5898,
+      "eval_samples_per_second": 469.621,
+      "eval_steps_per_second": 5.086,
+      "step": 80
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.7018,
+      "step": 100
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5270758122743683,
+      "eval_loss": 0.6921332478523254,
+      "eval_runtime": 0.5921,
+      "eval_samples_per_second": 467.809,
+      "eval_steps_per_second": 5.067,
+      "step": 100
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 3.52e-05,
+      "loss": 0.697,
+      "step": 120
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.4729241877256318,
+      "eval_loss": 0.7111871242523193,
+      "eval_runtime": 0.5887,
+      "eval_samples_per_second": 470.553,
+      "eval_steps_per_second": 5.096,
+      "step": 120
+    },
+    {
+      "epoch": 6.0,
+      "step": 120,
+      "total_flos": 2212565757198336.0,
+      "train_loss": 0.7045625448226929,
+      "train_runtime": 133.6091,
+      "train_samples_per_second": 931.822,
+      "train_steps_per_second": 7.485
+    }
+  ],
+  "max_steps": 1000,
+  "num_train_epochs": 50,
+  "total_flos": 2212565757198336.0,
+  "trial_name": null,
+  "trial_params": null
+}