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: 0.43661971830985913
+      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_new_pretrain_w_init__wnli
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_wt_init](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_wt_init) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_wt_init](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_wt_init) on the GLUE WNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7009
-- Accuracy: 0.4366
+- Loss: 0.6855
+- Accuracy: 0.5634
 
 ## Model description
 

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 14.0,
+    "eval_accuracy": 0.5633802816901409,
+    "eval_loss": 0.6855193376541138,
+    "eval_runtime": 0.1566,
+    "eval_samples": 71,
+    "eval_samples_per_second": 453.256,
+    "eval_steps_per_second": 6.384,
+    "train_loss": 2.0194963932037355,
+    "train_runtime": 90.6818,
+    "train_samples": 635,
+    "train_samples_per_second": 350.126,
+    "train_steps_per_second": 2.757
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 14.0,
+    "eval_accuracy": 0.5633802816901409,
+    "eval_loss": 0.6855193376541138,
+    "eval_runtime": 0.1566,
+    "eval_samples": 71,
+    "eval_samples_per_second": 453.256,
+    "eval_steps_per_second": 6.384
+}

logs/events.out.tfevents.1685552384.serv-9223.3032138.23

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dbb4f6845f72ab038d03ec97a08f558d671b351f91813f7c2b5bb1b82a06a5eb
+size 357

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 14.0,
+    "train_loss": 2.0194963932037355,
+    "train_runtime": 90.6818,
+    "train_samples": 635,
+    "train_samples_per_second": 350.126,
+    "train_steps_per_second": 2.757
+}

trainer_state.json

@@ -0,0 +1,235 @@
+{
+  "best_metric": 0.6855193376541138,
+  "best_model_checkpoint": "hBERTv1_new_pretrain_w_init__wnli/checkpoint-45",
+  "epoch": 14.0,
+  "global_step": 70,
+  "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": 12.3688,
+      "step": 5
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 6.223592281341553,
+      "eval_runtime": 0.1345,
+      "eval_samples_per_second": 527.809,
+      "eval_steps_per_second": 7.434,
+      "step": 5
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 0.00048,
+      "loss": 3.5093,
+      "step": 10
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 0.7491196393966675,
+      "eval_runtime": 0.1348,
+      "eval_samples_per_second": 526.695,
+      "eval_steps_per_second": 7.418,
+      "step": 10
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 0.00047,
+      "loss": 1.9112,
+      "step": 15
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 2.5145533084869385,
+      "eval_runtime": 0.1344,
+      "eval_samples_per_second": 528.182,
+      "eval_steps_per_second": 7.439,
+      "step": 15
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 0.00046,
+      "loss": 1.4995,
+      "step": 20
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 1.8103646039962769,
+      "eval_runtime": 0.1343,
+      "eval_samples_per_second": 528.656,
+      "eval_steps_per_second": 7.446,
+      "step": 20
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 0.00045000000000000004,
+      "loss": 1.3047,
+      "step": 25
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.6936343908309937,
+      "eval_runtime": 0.1346,
+      "eval_samples_per_second": 527.436,
+      "eval_steps_per_second": 7.429,
+      "step": 25
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 0.00044,
+      "loss": 1.4685,
+      "step": 30
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.7440030574798584,
+      "eval_runtime": 0.1342,
+      "eval_samples_per_second": 528.98,
+      "eval_steps_per_second": 7.45,
+      "step": 30
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 0.00043,
+      "loss": 0.924,
+      "step": 35
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 1.1065967082977295,
+      "eval_runtime": 0.1344,
+      "eval_samples_per_second": 528.272,
+      "eval_steps_per_second": 7.44,
+      "step": 35
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 0.00042,
+      "loss": 0.8423,
+      "step": 40
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 0.8221006989479065,
+      "eval_runtime": 0.1346,
+      "eval_samples_per_second": 527.644,
+      "eval_steps_per_second": 7.432,
+      "step": 40
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 0.00041,
+      "loss": 0.8166,
+      "step": 45
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.6855193376541138,
+      "eval_runtime": 0.1345,
+      "eval_samples_per_second": 528.041,
+      "eval_steps_per_second": 7.437,
+      "step": 45
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 0.0004,
+      "loss": 0.7552,
+      "step": 50
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.718089759349823,
+      "eval_runtime": 0.1345,
+      "eval_samples_per_second": 527.974,
+      "eval_steps_per_second": 7.436,
+      "step": 50
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 0.00039000000000000005,
+      "loss": 0.7515,
+      "step": 55
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.6951475739479065,
+      "eval_runtime": 0.1352,
+      "eval_samples_per_second": 525.152,
+      "eval_steps_per_second": 7.397,
+      "step": 55
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 0.00038,
+      "loss": 0.7127,
+      "step": 60
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 0.7139633893966675,
+      "eval_runtime": 0.1349,
+      "eval_samples_per_second": 526.347,
+      "eval_steps_per_second": 7.413,
+      "step": 60
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 0.00037,
+      "loss": 0.7112,
+      "step": 65
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.6901406645774841,
+      "eval_runtime": 0.1345,
+      "eval_samples_per_second": 527.704,
+      "eval_steps_per_second": 7.432,
+      "step": 65
+    },
+    {
+      "epoch": 14.0,
+      "learning_rate": 0.00035999999999999997,
+      "loss": 0.6976,
+      "step": 70
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 0.700924277305603,
+      "eval_runtime": 0.1352,
+      "eval_samples_per_second": 525.153,
+      "eval_steps_per_second": 7.397,
+      "step": 70
+    },
+    {
+      "epoch": 14.0,
+      "step": 70,
+      "total_flos": 1316580365434880.0,
+      "train_loss": 2.0194963932037355,
+      "train_runtime": 90.6818,
+      "train_samples_per_second": 350.126,
+      "train_steps_per_second": 2.757
+    }
+  ],
+  "max_steps": 250,
+  "num_train_epochs": 50,
+  "total_flos": 1316580365434880.0,
+  "trial_name": null,
+  "trial_params": null
+}