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: 0.49458483754512633
+      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. -->
 
 # hBERTv2_new_pretrain_48_KD_rte
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new_48_KD](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new_48_KD) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new_48_KD](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new_48_KD) on the GLUE RTE dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.1418
-- Accuracy: 0.4946
+- Loss: 0.6928
+- Accuracy: 0.5271
 
 ## Model description
 

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 6.0,
+    "eval_accuracy": 0.5270758122743683,
+    "eval_loss": 0.6927525401115417,
+    "eval_runtime": 0.6235,
+    "eval_samples": 277,
+    "eval_samples_per_second": 444.292,
+    "eval_steps_per_second": 4.812,
+    "train_loss": 0.6341694990793864,
+    "train_runtime": 132.2942,
+    "train_samples": 2490,
+    "train_samples_per_second": 941.084,
+    "train_steps_per_second": 7.559
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 6.0,
+    "eval_accuracy": 0.5270758122743683,
+    "eval_loss": 0.6927525401115417,
+    "eval_runtime": 0.6235,
+    "eval_samples": 277,
+    "eval_samples_per_second": 444.292,
+    "eval_steps_per_second": 4.812
+}

logs/events.out.tfevents.1686801279.serv-9207.3373475.11

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:366e1feea56824aa3f965dbff50fe86d9056253a336c82f03fc008c243ca28d7
+size 357

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 6.0,
+    "train_loss": 0.6341694990793864,
+    "train_runtime": 132.2942,
+    "train_samples": 2490,
+    "train_samples_per_second": 941.084,
+    "train_steps_per_second": 7.559
+}

trainer_state.json

@@ -0,0 +1,115 @@
+{
+  "best_metric": 0.6927525401115417,
+  "best_model_checkpoint": "hBERTv2_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": 0.7143,
+      "step": 20
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5270758122743683,
+      "eval_loss": 0.6927525401115417,
+      "eval_runtime": 0.6122,
+      "eval_samples_per_second": 452.497,
+      "eval_steps_per_second": 4.901,
+      "step": 20
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.6999,
+      "step": 40
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.49097472924187724,
+      "eval_loss": 0.6949225664138794,
+      "eval_runtime": 0.6063,
+      "eval_samples_per_second": 456.898,
+      "eval_steps_per_second": 4.948,
+      "step": 40
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.76e-05,
+      "loss": 0.7002,
+      "step": 60
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5451263537906137,
+      "eval_loss": 0.6943550109863281,
+      "eval_runtime": 0.6121,
+      "eval_samples_per_second": 452.521,
+      "eval_steps_per_second": 4.901,
+      "step": 60
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.6627,
+      "step": 80
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5090252707581228,
+      "eval_loss": 0.7554575800895691,
+      "eval_runtime": 0.6055,
+      "eval_samples_per_second": 457.506,
+      "eval_steps_per_second": 4.955,
+      "step": 80
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.5779,
+      "step": 100
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5018050541516246,
+      "eval_loss": 0.9458572864532471,
+      "eval_runtime": 0.6068,
+      "eval_samples_per_second": 456.529,
+      "eval_steps_per_second": 4.944,
+      "step": 100
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 3.52e-05,
+      "loss": 0.4501,
+      "step": 120
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.49458483754512633,
+      "eval_loss": 1.1417500972747803,
+      "eval_runtime": 0.6085,
+      "eval_samples_per_second": 455.238,
+      "eval_steps_per_second": 4.93,
+      "step": 120
+    },
+    {
+      "epoch": 6.0,
+      "step": 120,
+      "total_flos": 2186307132456960.0,
+      "train_loss": 0.6341694990793864,
+      "train_runtime": 132.2942,
+      "train_samples_per_second": 941.084,
+      "train_steps_per_second": 7.559
+    }
+  ],
+  "max_steps": 1000,
+  "num_train_epochs": 50,
+  "total_flos": 2186307132456960.0,
+  "trial_name": null,
+  "trial_params": null
+}