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 COLA
       type: glue
       config: cola
       split: validation
@@ -28,9 +30,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hBERTv2_cola
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2](https://huggingface.co/gokuls/bert_12_layer_model_v2) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2](https://huggingface.co/gokuls/bert_12_layer_model_v2) on the GLUE COLA dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6196
+- Loss: 0.6182
 - Matthews Correlation: 0.0
 
 ## Model description

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 7.0,
+    "eval_loss": 0.6181691884994507,
+    "eval_matthews_correlation": 0.0,
+    "eval_runtime": 1.2427,
+    "eval_samples": 1043,
+    "eval_samples_per_second": 839.315,
+    "eval_steps_per_second": 4.024,
+    "train_loss": 0.6130136441783745,
+    "train_runtime": 229.4882,
+    "train_samples": 8551,
+    "train_samples_per_second": 1863.059,
+    "train_steps_per_second": 7.408
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 7.0,
+    "eval_loss": 0.6181691884994507,
+    "eval_matthews_correlation": 0.0,
+    "eval_runtime": 1.2427,
+    "eval_samples": 1043,
+    "eval_samples_per_second": 839.315,
+    "eval_steps_per_second": 4.024
+}

logs/events.out.tfevents.1677660541.kyoto.1072791.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e6785d57ce1e1d0abcf0401ac1d43532386dab6e13793ebaf7a072e10666baf0
+size 375

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 7.0,
+    "train_loss": 0.6130136441783745,
+    "train_runtime": 229.4882,
+    "train_samples": 8551,
+    "train_samples_per_second": 1863.059,
+    "train_steps_per_second": 7.408
+}

trainer_state.json

@@ -0,0 +1,130 @@
+{
+  "best_metric": 0.6181691884994507,
+  "best_model_checkpoint": "hBERTv2_cola/checkpoint-68",
+  "epoch": 7.0,
+  "global_step": 238,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.9029411764705883e-05,
+      "loss": 0.6296,
+      "step": 34
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.618189811706543,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.2309,
+      "eval_samples_per_second": 847.325,
+      "eval_steps_per_second": 4.062,
+      "step": 34
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.8029411764705886e-05,
+      "loss": 0.614,
+      "step": 68
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.6181691884994507,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.2378,
+      "eval_samples_per_second": 842.642,
+      "eval_steps_per_second": 4.04,
+      "step": 68
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.702941176470588e-05,
+      "loss": 0.6089,
+      "step": 102
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.6213549971580505,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.2389,
+      "eval_samples_per_second": 841.871,
+      "eval_steps_per_second": 4.036,
+      "step": 102
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.6029411764705885e-05,
+      "loss": 0.6092,
+      "step": 136
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 0.6190556287765503,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.2383,
+      "eval_samples_per_second": 842.292,
+      "eval_steps_per_second": 4.038,
+      "step": 136
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.502941176470589e-05,
+      "loss": 0.6103,
+      "step": 170
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 0.6221293210983276,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.2416,
+      "eval_samples_per_second": 840.076,
+      "eval_steps_per_second": 4.027,
+      "step": 170
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4029411764705884e-05,
+      "loss": 0.6094,
+      "step": 204
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 0.6182675361633301,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.2396,
+      "eval_samples_per_second": 841.387,
+      "eval_steps_per_second": 4.033,
+      "step": 204
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.302941176470589e-05,
+      "loss": 0.6096,
+      "step": 238
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 0.6196076273918152,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.2395,
+      "eval_samples_per_second": 841.485,
+      "eval_steps_per_second": 4.034,
+      "step": 238
+    },
+    {
+      "epoch": 7.0,
+      "step": 238,
+      "total_flos": 7455394754461696.0,
+      "train_loss": 0.6130136441783745,
+      "train_runtime": 229.4882,
+      "train_samples_per_second": 1863.059,
+      "train_steps_per_second": 7.408
+    }
+  ],
+  "max_steps": 1700,
+  "num_train_epochs": 50,
+  "total_flos": 7455394754461696.0,
+  "trial_name": null,
+  "trial_params": null
+}