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. -->
 
 # hBERTv1_new_pretrain_cola
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new) on the GLUE COLA dataset.
 It achieves the following results on the evaluation set:
-- Loss: 16.5733
+- Loss: 16.5460
 - Matthews Correlation: 0.0
 - Accuracy: 0.6913
 

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.6912751793861389,
+    "eval_loss": 16.5460205078125,
+    "eval_matthews_correlation": 0.0,
+    "eval_runtime": 1.8993,
+    "eval_samples": 1043,
+    "eval_samples_per_second": 549.142,
+    "eval_steps_per_second": 4.739,
+    "train_loss": 17.391823156556086,
+    "train_runtime": 416.7234,
+    "train_samples": 8551,
+    "train_samples_per_second": 1025.98,
+    "train_steps_per_second": 8.039
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.6912751793861389,
+    "eval_loss": 16.5460205078125,
+    "eval_matthews_correlation": 0.0,
+    "eval_runtime": 1.8993,
+    "eval_samples": 1043,
+    "eval_samples_per_second": 549.142,
+    "eval_steps_per_second": 4.739
+}

logs/events.out.tfevents.1685530821.serv-3318.2597307.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d766b497be7d56bfa5e401d6878a29aeb1de4928a32b2330d2e2400f80154dc4
+size 427

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 8.0,
+    "train_loss": 17.391823156556086,
+    "train_runtime": 416.7234,
+    "train_samples": 8551,
+    "train_samples_per_second": 1025.98,
+    "train_steps_per_second": 8.039
+}

trainer_state.json

@@ -0,0 +1,153 @@
+{
+  "best_metric": 16.5460205078125,
+  "best_model_checkpoint": "hBERTv1_new_pretrain_cola/checkpoint-201",
+  "epoch": 8.0,
+  "global_step": 536,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.00049,
+      "loss": 17.7846,
+      "step": 67
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6912751793861389,
+      "eval_loss": 16.576702117919922,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.9107,
+      "eval_samples_per_second": 545.874,
+      "eval_steps_per_second": 4.71,
+      "step": 67
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 0.00048,
+      "loss": 17.5238,
+      "step": 134
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6912751793861389,
+      "eval_loss": 16.567354202270508,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.9081,
+      "eval_samples_per_second": 546.627,
+      "eval_steps_per_second": 4.717,
+      "step": 134
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 0.00047,
+      "loss": 17.4369,
+      "step": 201
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6912751793861389,
+      "eval_loss": 16.5460205078125,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.9122,
+      "eval_samples_per_second": 545.459,
+      "eval_steps_per_second": 4.707,
+      "step": 201
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 0.00046,
+      "loss": 17.0194,
+      "step": 268
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6912751793861389,
+      "eval_loss": 16.574783325195312,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.911,
+      "eval_samples_per_second": 545.784,
+      "eval_steps_per_second": 4.71,
+      "step": 268
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 0.00045000000000000004,
+      "loss": 17.1371,
+      "step": 335
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6912751793861389,
+      "eval_loss": 16.56974983215332,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.9125,
+      "eval_samples_per_second": 545.364,
+      "eval_steps_per_second": 4.706,
+      "step": 335
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 0.00044,
+      "loss": 17.4193,
+      "step": 402
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6912751793861389,
+      "eval_loss": 16.57382583618164,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.9111,
+      "eval_samples_per_second": 545.751,
+      "eval_steps_per_second": 4.709,
+      "step": 402
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 0.00043,
+      "loss": 17.2534,
+      "step": 469
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6912751793861389,
+      "eval_loss": 16.577661514282227,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.9094,
+      "eval_samples_per_second": 546.245,
+      "eval_steps_per_second": 4.714,
+      "step": 469
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 0.00042,
+      "loss": 17.56,
+      "step": 536
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6912751793861389,
+      "eval_loss": 16.573345184326172,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.9063,
+      "eval_samples_per_second": 547.146,
+      "eval_steps_per_second": 4.721,
+      "step": 536
+    },
+    {
+      "epoch": 8.0,
+      "step": 536,
+      "total_flos": 1.0131004291284992e+16,
+      "train_loss": 17.391823156556086,
+      "train_runtime": 416.7234,
+      "train_samples_per_second": 1025.98,
+      "train_steps_per_second": 8.039
+    }
+  ],
+  "max_steps": 3350,
+  "num_train_epochs": 50,
+  "total_flos": 1.0131004291284992e+16,
+  "trial_name": null,
+  "trial_params": null
+}