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 SST2
       type: glue
       config: sst2
       split: validation
@@ -28,9 +30,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hBERTv2_sst2
 
-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 SST2 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6980
+- Loss: 0.6964
 - Accuracy: 0.5092
 
 ## Model description

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 14.0,
+    "eval_accuracy": 0.5091743119266054,
+    "eval_loss": 0.6963669061660767,
+    "eval_runtime": 1.0895,
+    "eval_samples": 872,
+    "eval_samples_per_second": 800.333,
+    "eval_steps_per_second": 3.671,
+    "train_loss": 0.6872374216715494,
+    "train_runtime": 3475.6313,
+    "train_samples": 67349,
+    "train_samples_per_second": 968.874,
+    "train_steps_per_second": 3.798
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 14.0,
+    "eval_accuracy": 0.5091743119266054,
+    "eval_loss": 0.6963669061660767,
+    "eval_runtime": 1.0895,
+    "eval_samples": 872,
+    "eval_samples_per_second": 800.333,
+    "eval_steps_per_second": 3.671
+}

logs/events.out.tfevents.1677681175.kyoto.1072791.17

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:25d72ab7b7fc34d6fb985f0586c058dcdc348452512e38b81a456f888ce7f77d
+size 363

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 14.0,
+    "train_loss": 0.6872374216715494,
+    "train_runtime": 3475.6313,
+    "train_samples": 67349,
+    "train_samples_per_second": 968.874,
+    "train_steps_per_second": 3.798
+}

trainer_state.json

@@ -0,0 +1,235 @@
+{
+  "best_metric": 0.6963669061660767,
+  "best_model_checkpoint": "hBERTv2_sst2/checkpoint-2376",
+  "epoch": 14.0,
+  "global_step": 3696,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.9003787878787885e-05,
+      "loss": 0.6916,
+      "step": 264
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5091743119266054,
+      "eval_loss": 0.6998637914657593,
+      "eval_runtime": 1.0361,
+      "eval_samples_per_second": 841.637,
+      "eval_steps_per_second": 3.861,
+      "step": 264
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.800378787878788e-05,
+      "loss": 0.6885,
+      "step": 528
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5091743119266054,
+      "eval_loss": 0.6977964639663696,
+      "eval_runtime": 1.037,
+      "eval_samples_per_second": 840.878,
+      "eval_steps_per_second": 3.857,
+      "step": 528
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7003787878787884e-05,
+      "loss": 0.6871,
+      "step": 792
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5091743119266054,
+      "eval_loss": 0.698425829410553,
+      "eval_runtime": 1.0359,
+      "eval_samples_per_second": 841.809,
+      "eval_steps_per_second": 3.862,
+      "step": 792
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.600378787878788e-05,
+      "loss": 0.6869,
+      "step": 1056
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5091743119266054,
+      "eval_loss": 0.69902503490448,
+      "eval_runtime": 1.0372,
+      "eval_samples_per_second": 840.699,
+      "eval_steps_per_second": 3.856,
+      "step": 1056
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.5003787878787876e-05,
+      "loss": 0.6868,
+      "step": 1320
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5091743119266054,
+      "eval_loss": 0.6974067091941833,
+      "eval_runtime": 1.0355,
+      "eval_samples_per_second": 842.083,
+      "eval_steps_per_second": 3.863,
+      "step": 1320
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4003787878787885e-05,
+      "loss": 0.6869,
+      "step": 1584
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5091743119266054,
+      "eval_loss": 0.6980025172233582,
+      "eval_runtime": 1.0358,
+      "eval_samples_per_second": 841.851,
+      "eval_steps_per_second": 3.862,
+      "step": 1584
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.300378787878788e-05,
+      "loss": 0.6867,
+      "step": 1848
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5091743119266054,
+      "eval_loss": 0.6984168887138367,
+      "eval_runtime": 1.0568,
+      "eval_samples_per_second": 825.136,
+      "eval_steps_per_second": 3.785,
+      "step": 1848
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.2003787878787884e-05,
+      "loss": 0.6868,
+      "step": 2112
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5091743119266054,
+      "eval_loss": 0.6974700093269348,
+      "eval_runtime": 1.0448,
+      "eval_samples_per_second": 834.6,
+      "eval_steps_per_second": 3.828,
+      "step": 2112
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 4.100378787878788e-05,
+      "loss": 0.6868,
+      "step": 2376
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.5091743119266054,
+      "eval_loss": 0.6963669061660767,
+      "eval_runtime": 1.037,
+      "eval_samples_per_second": 840.85,
+      "eval_steps_per_second": 3.857,
+      "step": 2376
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 4.0003787878787876e-05,
+      "loss": 0.6865,
+      "step": 2640
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.5091743119266054,
+      "eval_loss": 0.6977964639663696,
+      "eval_runtime": 1.0355,
+      "eval_samples_per_second": 842.138,
+      "eval_steps_per_second": 3.863,
+      "step": 2640
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 3.900378787878788e-05,
+      "loss": 0.6868,
+      "step": 2904
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.5091743119266054,
+      "eval_loss": 0.6980271935462952,
+      "eval_runtime": 1.0325,
+      "eval_samples_per_second": 844.563,
+      "eval_steps_per_second": 3.874,
+      "step": 2904
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 3.800378787878788e-05,
+      "loss": 0.6865,
+      "step": 3168
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.5091743119266054,
+      "eval_loss": 0.700083315372467,
+      "eval_runtime": 1.0347,
+      "eval_samples_per_second": 842.747,
+      "eval_steps_per_second": 3.866,
+      "step": 3168
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 3.7003787878787885e-05,
+      "loss": 0.6867,
+      "step": 3432
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.5091743119266054,
+      "eval_loss": 0.6965734958648682,
+      "eval_runtime": 1.032,
+      "eval_samples_per_second": 844.972,
+      "eval_steps_per_second": 3.876,
+      "step": 3432
+    },
+    {
+      "epoch": 14.0,
+      "learning_rate": 3.600378787878788e-05,
+      "loss": 0.6867,
+      "step": 3696
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.5091743119266054,
+      "eval_loss": 0.6980003118515015,
+      "eval_runtime": 1.0332,
+      "eval_samples_per_second": 843.945,
+      "eval_steps_per_second": 3.871,
+      "step": 3696
+    },
+    {
+      "epoch": 14.0,
+      "step": 3696,
+      "total_flos": 1.1743968521879552e+17,
+      "train_loss": 0.6872374216715494,
+      "train_runtime": 3475.6313,
+      "train_samples_per_second": 968.874,
+      "train_steps_per_second": 3.798
+    }
+  ],
+  "max_steps": 13200,
+  "num_train_epochs": 50,
+  "total_flos": 1.1743968521879552e+17,
+  "trial_name": null,
+  "trial_params": null
+}