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 WNLI
       type: glue
       config: wnli
       split: validation
@@ -20,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.5211267605633803
+      value: 0.5633802816901409
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -28,10 +30,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hBERTv1_wnli
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1](https://huggingface.co/gokuls/bert_12_layer_model_v1) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1](https://huggingface.co/gokuls/bert_12_layer_model_v1) on the GLUE WNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6925
-- Accuracy: 0.5211
+- Loss: 0.6877
+- Accuracy: 0.5634
 
 ## Model description
 

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 17.0,
+    "eval_accuracy": 0.5633802816901409,
+    "eval_loss": 0.687747597694397,
+    "eval_runtime": 0.1167,
+    "eval_samples": 71,
+    "eval_samples_per_second": 608.336,
+    "eval_steps_per_second": 8.568,
+    "train_loss": 0.6992356636944939,
+    "train_runtime": 99.2747,
+    "train_samples": 635,
+    "train_samples_per_second": 319.82,
+    "train_steps_per_second": 1.511
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 17.0,
+    "eval_accuracy": 0.5633802816901409,
+    "eval_loss": 0.687747597694397,
+    "eval_runtime": 0.1167,
+    "eval_samples": 71,
+    "eval_samples_per_second": 608.336,
+    "eval_steps_per_second": 8.568
+}

logs/events.out.tfevents.1677756438.serv-9223.249682.23

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76524319446a238c0c6d5870e5c43c8749f4868a7749087044db8494b8838b2e
+size 357

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 17.0,
+    "train_loss": 0.6992356636944939,
+    "train_runtime": 99.2747,
+    "train_samples": 635,
+    "train_samples_per_second": 319.82,
+    "train_steps_per_second": 1.511
+}

trainer_state.json

@@ -0,0 +1,280 @@
+{
+  "best_metric": 0.687747597694397,
+  "best_model_checkpoint": "hBERTv1_wnli/checkpoint-36",
+  "epoch": 17.0,
+  "global_step": 51,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.9e-05,
+      "loss": 0.7359,
+      "step": 3
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 0.7193758487701416,
+      "eval_runtime": 0.1035,
+      "eval_samples_per_second": 685.963,
+      "eval_steps_per_second": 9.661,
+      "step": 3
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.8e-05,
+      "loss": 0.6989,
+      "step": 6
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.68994140625,
+      "eval_runtime": 0.0997,
+      "eval_samples_per_second": 712.235,
+      "eval_steps_per_second": 10.031,
+      "step": 6
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7e-05,
+      "loss": 0.7031,
+      "step": 9
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 0.7027672529220581,
+      "eval_runtime": 0.0992,
+      "eval_samples_per_second": 715.838,
+      "eval_steps_per_second": 10.082,
+      "step": 9
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.7012,
+      "step": 12
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.6888756155967712,
+      "eval_runtime": 0.0988,
+      "eval_samples_per_second": 718.868,
+      "eval_steps_per_second": 10.125,
+      "step": 12
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.5e-05,
+      "loss": 0.697,
+      "step": 15
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.6893500089645386,
+      "eval_runtime": 0.0984,
+      "eval_samples_per_second": 721.2,
+      "eval_steps_per_second": 10.158,
+      "step": 15
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.6971,
+      "step": 18
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 0.701467752456665,
+      "eval_runtime": 0.0981,
+      "eval_samples_per_second": 723.896,
+      "eval_steps_per_second": 10.196,
+      "step": 18
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.3e-05,
+      "loss": 0.7,
+      "step": 21
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.6881740093231201,
+      "eval_runtime": 0.1052,
+      "eval_samples_per_second": 674.863,
+      "eval_steps_per_second": 9.505,
+      "step": 21
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.2e-05,
+      "loss": 0.6928,
+      "step": 24
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.6890406012535095,
+      "eval_runtime": 0.1,
+      "eval_samples_per_second": 710.295,
+      "eval_steps_per_second": 10.004,
+      "step": 24
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 4.1e-05,
+      "loss": 0.6932,
+      "step": 27
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.689693808555603,
+      "eval_runtime": 0.0997,
+      "eval_samples_per_second": 712.293,
+      "eval_steps_per_second": 10.032,
+      "step": 27
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 4e-05,
+      "loss": 0.6954,
+      "step": 30
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 0.695560097694397,
+      "eval_runtime": 0.1007,
+      "eval_samples_per_second": 704.825,
+      "eval_steps_per_second": 9.927,
+      "step": 30
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 3.9000000000000006e-05,
+      "loss": 0.6962,
+      "step": 33
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.6913100481033325,
+      "eval_runtime": 0.0999,
+      "eval_samples_per_second": 710.909,
+      "eval_steps_per_second": 10.013,
+      "step": 33
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 3.8e-05,
+      "loss": 0.6956,
+      "step": 36
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.687747597694397,
+      "eval_runtime": 0.1,
+      "eval_samples_per_second": 710.233,
+      "eval_steps_per_second": 10.003,
+      "step": 36
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 3.7e-05,
+      "loss": 0.6973,
+      "step": 39
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.5070422535211268,
+      "eval_loss": 0.6925820708274841,
+      "eval_runtime": 0.1034,
+      "eval_samples_per_second": 686.38,
+      "eval_steps_per_second": 9.667,
+      "step": 39
+    },
+    {
+      "epoch": 14.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.6978,
+      "step": 42
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.49295774647887325,
+      "eval_loss": 0.6933045983314514,
+      "eval_runtime": 0.1003,
+      "eval_samples_per_second": 707.79,
+      "eval_steps_per_second": 9.969,
+      "step": 42
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 3.5e-05,
+      "loss": 0.6945,
+      "step": 45
+    },
+    {
+      "epoch": 15.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.688297688961029,
+      "eval_runtime": 0.0985,
+      "eval_samples_per_second": 720.537,
+      "eval_steps_per_second": 10.148,
+      "step": 45
+    },
+    {
+      "epoch": 16.0,
+      "learning_rate": 3.4000000000000007e-05,
+      "loss": 0.6974,
+      "step": 48
+    },
+    {
+      "epoch": 16.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.6881258487701416,
+      "eval_runtime": 0.0996,
+      "eval_samples_per_second": 712.95,
+      "eval_steps_per_second": 10.042,
+      "step": 48
+    },
+    {
+      "epoch": 17.0,
+      "learning_rate": 3.3e-05,
+      "loss": 0.6936,
+      "step": 51
+    },
+    {
+      "epoch": 17.0,
+      "eval_accuracy": 0.5211267605633803,
+      "eval_loss": 0.6924790143966675,
+      "eval_runtime": 0.0992,
+      "eval_samples_per_second": 715.872,
+      "eval_steps_per_second": 10.083,
+      "step": 51
+    },
+    {
+      "epoch": 17.0,
+      "step": 51,
+      "total_flos": 1363527629537280.0,
+      "train_loss": 0.6992356636944939,
+      "train_runtime": 99.2747,
+      "train_samples_per_second": 319.82,
+      "train_steps_per_second": 1.511
+    }
+  ],
+  "max_steps": 150,
+  "num_train_epochs": 50,
+  "total_flos": 1363527629537280.0,
+  "trial_name": null,
+  "trial_params": null
+}