End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 base_model: gokuls/bert_12_layer_model_v1_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 RTE
       type: glue
       config: rte
       split: validation
@@ -21,7 +23,7 @@ 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
@@ -29,10 +31,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hBERTv1_new_pretrain_w_init_48_ver2_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: 0.7078
-- Accuracy: 0.4729
+- Loss: 0.6919
+- Accuracy: 0.5271
 
 ## Model description
 

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 6.0,
+    "eval_accuracy": 0.5270758122743683,
+    "eval_loss": 0.691931962966919,
+    "eval_runtime": 1.4981,
+    "eval_samples": 277,
+    "eval_samples_per_second": 184.901,
+    "eval_steps_per_second": 3.338,
+    "train_loss": 0.7067025421012161,
+    "train_runtime": 264.6778,
+    "train_samples": 2490,
+    "train_samples_per_second": 141.115,
+    "train_steps_per_second": 2.21
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 6.0,
+    "eval_accuracy": 0.5270758122743683,
+    "eval_loss": 0.691931962966919,
+    "eval_runtime": 1.4981,
+    "eval_samples": 277,
+    "eval_samples_per_second": 184.901,
+    "eval_steps_per_second": 3.338
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 6.0,
+    "train_loss": 0.7067025421012161,
+    "train_runtime": 264.6778,
+    "train_samples": 2490,
+    "train_samples_per_second": 141.115,
+    "train_steps_per_second": 2.21
+}

trainer_state.json

@@ -0,0 +1,118 @@
+{
+  "best_metric": 0.691931962966919,
+  "best_model_checkpoint": "hBERTv1_new_pretrain_w_init_48_ver2_rte/checkpoint-39",
+  "epoch": 6.0,
+  "eval_steps": 500,
+  "global_step": 234,
+  "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": 0.7175,
+      "step": 39
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5270758122743683,
+      "eval_loss": 0.691931962966919,
+      "eval_runtime": 1.4647,
+      "eval_samples_per_second": 189.113,
+      "eval_steps_per_second": 3.414,
+      "step": 39
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.466666666666667e-05,
+      "loss": 0.7099,
+      "step": 78
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.4729241877256318,
+      "eval_loss": 0.7006047368049622,
+      "eval_runtime": 1.4696,
+      "eval_samples_per_second": 188.487,
+      "eval_steps_per_second": 3.402,
+      "step": 78
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.7111,
+      "step": 117
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5270758122743683,
+      "eval_loss": 0.6926825046539307,
+      "eval_runtime": 1.4718,
+      "eval_samples_per_second": 188.209,
+      "eval_steps_per_second": 3.397,
+      "step": 117
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 2.9333333333333333e-05,
+      "loss": 0.7011,
+      "step": 156
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.4729241877256318,
+      "eval_loss": 0.6976003050804138,
+      "eval_runtime": 1.4634,
+      "eval_samples_per_second": 189.29,
+      "eval_steps_per_second": 3.417,
+      "step": 156
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 2.6666666666666667e-05,
+      "loss": 0.7021,
+      "step": 195
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.4729241877256318,
+      "eval_loss": 0.695507824420929,
+      "eval_runtime": 1.4648,
+      "eval_samples_per_second": 189.108,
+      "eval_steps_per_second": 3.414,
+      "step": 195
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 2.4e-05,
+      "loss": 0.6986,
+      "step": 234
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.4729241877256318,
+      "eval_loss": 0.7077711820602417,
+      "eval_runtime": 1.4767,
+      "eval_samples_per_second": 187.58,
+      "eval_steps_per_second": 3.386,
+      "step": 234
+    },
+    {
+      "epoch": 6.0,
+      "step": 234,
+      "total_flos": 2212565757198336.0,
+      "train_loss": 0.7067025421012161,
+      "train_runtime": 264.6778,
+      "train_samples_per_second": 141.115,
+      "train_steps_per_second": 2.21
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 585,
+  "num_train_epochs": 15,
+  "save_steps": 500,
+  "total_flos": 2212565757198336.0,
+  "trial_name": null,
+  "trial_params": null
+}