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: 0.5090252707581228
+      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_w_init_48_rte
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_48](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_48) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_48](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_48) on the GLUE RTE dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.8002
-- Accuracy: 0.5090
+- Loss: 0.6910
+- Accuracy: 0.5271
 
 ## Model description
 

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 14.0,
+    "eval_accuracy": 0.5270758122743683,
+    "eval_loss": 0.6909549832344055,
+    "eval_runtime": 0.5562,
+    "eval_samples": 277,
+    "eval_samples_per_second": 498.054,
+    "eval_steps_per_second": 5.394,
+    "train_loss": 0.6911461523600987,
+    "train_runtime": 271.0754,
+    "train_samples": 2490,
+    "train_samples_per_second": 459.282,
+    "train_steps_per_second": 3.689
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 14.0,
+    "eval_accuracy": 0.5270758122743683,
+    "eval_loss": 0.6909549832344055,
+    "eval_runtime": 0.5562,
+    "eval_samples": 277,
+    "eval_samples_per_second": 498.054,
+    "eval_steps_per_second": 5.394
+}

logs/events.out.tfevents.1686057798.serv-9207.2389173.17

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df0ab267d246595837c93d5a9bc8b99996da040dfcff11d6a6d33156f60f97e4
+size 363

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 14.0,
+    "train_loss": 0.6911461523600987,
+    "train_runtime": 271.0754,
+    "train_samples": 2490,
+    "train_samples_per_second": 459.282,
+    "train_steps_per_second": 3.689
+}

trainer_state.json

@@ -0,0 +1,235 @@
+{
+  "best_metric": 0.6909549832344055,
+  "best_model_checkpoint": "hBERTv1_new_pretrain_w_init_48_rte/checkpoint-180",
+  "epoch": 14.0,
+  "global_step": 280,
+  "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": 0.712,
+      "step": 20
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.4729241877256318,
+      "eval_loss": 0.7158167958259583,
+      "eval_runtime": 0.5268,
+      "eval_samples_per_second": 525.771,
+      "eval_steps_per_second": 5.694,
+      "step": 20
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.7094,
+      "step": 40
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.4729241877256318,
+      "eval_loss": 0.6957778930664062,
+      "eval_runtime": 0.5296,
+      "eval_samples_per_second": 523.049,
+      "eval_steps_per_second": 5.665,
+      "step": 40
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.76e-05,
+      "loss": 0.7025,
+      "step": 60
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.4729241877256318,
+      "eval_loss": 0.7007981538772583,
+      "eval_runtime": 0.5287,
+      "eval_samples_per_second": 523.935,
+      "eval_steps_per_second": 5.674,
+      "step": 60
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.705,
+      "step": 80
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5270758122743683,
+      "eval_loss": 0.691913902759552,
+      "eval_runtime": 0.5289,
+      "eval_samples_per_second": 523.717,
+      "eval_steps_per_second": 5.672,
+      "step": 80
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.7023,
+      "step": 100
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5270758122743683,
+      "eval_loss": 0.6959893703460693,
+      "eval_runtime": 0.53,
+      "eval_samples_per_second": 522.685,
+      "eval_steps_per_second": 5.661,
+      "step": 100
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 3.52e-05,
+      "loss": 0.7002,
+      "step": 120
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.4729241877256318,
+      "eval_loss": 0.7094991207122803,
+      "eval_runtime": 0.5279,
+      "eval_samples_per_second": 524.689,
+      "eval_steps_per_second": 5.683,
+      "step": 120
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 3.44e-05,
+      "loss": 0.7071,
+      "step": 140
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.4729241877256318,
+      "eval_loss": 0.704041600227356,
+      "eval_runtime": 0.5311,
+      "eval_samples_per_second": 521.544,
+      "eval_steps_per_second": 5.648,
+      "step": 140
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 0.6982,
+      "step": 160
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5270758122743683,
+      "eval_loss": 0.6917728781700134,
+      "eval_runtime": 0.5338,
+      "eval_samples_per_second": 518.933,
+      "eval_steps_per_second": 5.62,
+      "step": 160
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 3.28e-05,
+      "loss": 0.7025,
+      "step": 180
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.5270758122743683,
+      "eval_loss": 0.6909549832344055,
+      "eval_runtime": 0.5302,
+      "eval_samples_per_second": 522.469,
+      "eval_steps_per_second": 5.659,
+      "step": 180
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.6965,
+      "step": 200
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.4620938628158845,
+      "eval_loss": 0.6983726024627686,
+      "eval_runtime": 0.5283,
+      "eval_samples_per_second": 524.331,
+      "eval_steps_per_second": 5.679,
+      "step": 200
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 3.1200000000000006e-05,
+      "loss": 0.6814,
+      "step": 220
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.49458483754512633,
+      "eval_loss": 0.7635096907615662,
+      "eval_runtime": 0.5287,
+      "eval_samples_per_second": 523.93,
+      "eval_steps_per_second": 5.674,
+      "step": 220
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 3.0400000000000004e-05,
+      "loss": 0.6616,
+      "step": 240
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.5270758122743683,
+      "eval_loss": 0.6918363571166992,
+      "eval_runtime": 0.5312,
+      "eval_samples_per_second": 521.504,
+      "eval_steps_per_second": 5.648,
+      "step": 240
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 2.96e-05,
+      "loss": 0.6658,
+      "step": 260
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.5306859205776173,
+      "eval_loss": 0.7622052431106567,
+      "eval_runtime": 0.5281,
+      "eval_samples_per_second": 524.555,
+      "eval_steps_per_second": 5.681,
+      "step": 260
+    },
+    {
+      "epoch": 14.0,
+      "learning_rate": 2.8800000000000002e-05,
+      "loss": 0.6316,
+      "step": 280
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.5090252707581228,
+      "eval_loss": 0.8002101182937622,
+      "eval_runtime": 0.5288,
+      "eval_samples_per_second": 523.798,
+      "eval_steps_per_second": 5.673,
+      "step": 280
+    },
+    {
+      "epoch": 14.0,
+      "step": 280,
+      "total_flos": 5162653433462784.0,
+      "train_loss": 0.6911461523600987,
+      "train_runtime": 271.0754,
+      "train_samples_per_second": 459.282,
+      "train_steps_per_second": 3.689
+    }
+  ],
+  "max_steps": 1000,
+  "num_train_epochs": 50,
+  "total_flos": 5162653433462784.0,
+  "trial_name": null,
+  "trial_params": null
+}