End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 tags:
 - generated_from_trainer
 datasets:
@@ -13,7 +15,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE COLA
       type: glue
       config: cola
       split: validation
@@ -32,9 +34,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hBERTv2_new_pretrain_w_init__cola
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new_wt_init](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new_wt_init) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new_wt_init](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new_wt_init) on the GLUE COLA dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6192
+- Loss: 0.6176
 - Matthews Correlation: 0.0
 - Accuracy: 0.6913
 

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.6912751793861389,
+    "eval_loss": 0.6176144480705261,
+    "eval_matthews_correlation": 0.0,
+    "eval_runtime": 1.7886,
+    "eval_samples": 1043,
+    "eval_samples_per_second": 583.147,
+    "eval_steps_per_second": 5.032,
+    "train_loss": 0.6147046770368304,
+    "train_runtime": 352.0595,
+    "train_samples": 8551,
+    "train_samples_per_second": 1214.425,
+    "train_steps_per_second": 9.515
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.6912751793861389,
+    "eval_loss": 0.6176144480705261,
+    "eval_matthews_correlation": 0.0,
+    "eval_runtime": 1.7886,
+    "eval_samples": 1043,
+    "eval_samples_per_second": 583.147,
+    "eval_steps_per_second": 5.032
+}

logs/events.out.tfevents.1686033119.serv-3317.3380349.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:00a3dc6c265cbfb08a395ffc74e9612940f913d0f3e8b9b63d617e420b628d6c
+size 427

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 7.0,
+    "train_loss": 0.6147046770368304,
+    "train_runtime": 352.0595,
+    "train_samples": 8551,
+    "train_samples_per_second": 1214.425,
+    "train_steps_per_second": 9.515
+}

trainer_state.json

@@ -0,0 +1,137 @@
+{
+  "best_metric": 0.6176144480705261,
+  "best_model_checkpoint": "hBERTv2_new_pretrain_w_init__cola/checkpoint-134",
+  "epoch": 7.0,
+  "global_step": 469,
+  "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": 0.6224,
+      "step": 67
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6912751793861389,
+      "eval_loss": 0.6208990216255188,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.7938,
+      "eval_samples_per_second": 581.453,
+      "eval_steps_per_second": 5.017,
+      "step": 67
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.6167,
+      "step": 134
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6912751793861389,
+      "eval_loss": 0.6176144480705261,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.7974,
+      "eval_samples_per_second": 580.267,
+      "eval_steps_per_second": 5.007,
+      "step": 134
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.76e-05,
+      "loss": 0.6138,
+      "step": 201
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6912751793861389,
+      "eval_loss": 0.6233745813369751,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.7973,
+      "eval_samples_per_second": 580.302,
+      "eval_steps_per_second": 5.007,
+      "step": 201
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.6162,
+      "step": 268
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6912751793861389,
+      "eval_loss": 0.6188260316848755,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.7966,
+      "eval_samples_per_second": 580.545,
+      "eval_steps_per_second": 5.009,
+      "step": 268
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.6108,
+      "step": 335
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6912751793861389,
+      "eval_loss": 0.620170533657074,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.7982,
+      "eval_samples_per_second": 580.021,
+      "eval_steps_per_second": 5.005,
+      "step": 335
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 3.52e-05,
+      "loss": 0.6114,
+      "step": 402
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6912751793861389,
+      "eval_loss": 0.6180151700973511,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.7959,
+      "eval_samples_per_second": 580.759,
+      "eval_steps_per_second": 5.011,
+      "step": 402
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 3.44e-05,
+      "loss": 0.6117,
+      "step": 469
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6912751793861389,
+      "eval_loss": 0.619239866733551,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.798,
+      "eval_samples_per_second": 580.089,
+      "eval_steps_per_second": 5.006,
+      "step": 469
+    },
+    {
+      "epoch": 7.0,
+      "step": 469,
+      "total_flos": 8759423665176576.0,
+      "train_loss": 0.6147046770368304,
+      "train_runtime": 352.0595,
+      "train_samples_per_second": 1214.425,
+      "train_steps_per_second": 9.515
+    }
+  ],
+  "max_steps": 3350,
+  "num_train_epochs": 50,
+  "total_flos": 8759423665176576.0,
+  "trial_name": null,
+  "trial_params": null
+}