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_48_emb_com_wnli
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_emb_compress_48](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_emb_compress_48) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_emb_compress_48](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_emb_compress_48) on the GLUE WNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7185
-- Accuracy: 0.4366
+- Loss: 0.6859
+- Accuracy: 0.5634
 
 ## Model description
 

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 9.0,
+    "eval_accuracy": 0.5633802816901409,
+    "eval_loss": 0.6859394311904907,
+    "eval_runtime": 0.3595,
+    "eval_samples": 71,
+    "eval_samples_per_second": 197.521,
+    "eval_steps_per_second": 2.782,
+    "train_loss": 0.7311807526482477,
+    "train_runtime": 109.9802,
+    "train_samples": 635,
+    "train_samples_per_second": 288.688,
+    "train_steps_per_second": 2.273
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 9.0,
+    "eval_accuracy": 0.5633802816901409,
+    "eval_loss": 0.6859394311904907,
+    "eval_runtime": 0.3595,
+    "eval_samples": 71,
+    "eval_samples_per_second": 197.521,
+    "eval_steps_per_second": 2.782
+}

logs/events.out.tfevents.1686837332.garda.2601621.15

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46f49a69ccb55b92722c03ebc00bd225edd4f70c31db72aca841e18a01ea2f86
+size 357

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 9.0,
+    "train_loss": 0.7311807526482477,
+    "train_runtime": 109.9802,
+    "train_samples": 635,
+    "train_samples_per_second": 288.688,
+    "train_steps_per_second": 2.273
+}

trainer_state.json

@@ -0,0 +1,160 @@
+{
+  "best_metric": 0.6859394311904907,
+  "best_model_checkpoint": "hBERTv1_new_pretrain_48_emb_com_wnli/checkpoint-20",
+  "epoch": 9.0,
+  "global_step": 45,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9200000000000004e-05,
+      "loss": 0.8985,
+      "step": 5
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 0.9143850803375244,
+      "eval_runtime": 0.3482,
+      "eval_samples_per_second": 203.885,
+      "eval_steps_per_second": 2.872,
+      "step": 5
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.7419,
+      "step": 10
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 0.7704455852508545,
+      "eval_runtime": 0.3494,
+      "eval_samples_per_second": 203.201,
+      "eval_steps_per_second": 2.862,
+      "step": 10
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.76e-05,
+      "loss": 0.7079,
+      "step": 15
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 0.7120934724807739,
+      "eval_runtime": 0.3499,
+      "eval_samples_per_second": 202.912,
+      "eval_steps_per_second": 2.858,
+      "step": 15
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.6978,
+      "step": 20
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.6859394311904907,
+      "eval_runtime": 0.3494,
+      "eval_samples_per_second": 203.222,
+      "eval_steps_per_second": 2.862,
+      "step": 20
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.7001,
+      "step": 25
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 0.7479454874992371,
+      "eval_runtime": 0.3493,
+      "eval_samples_per_second": 203.242,
+      "eval_steps_per_second": 2.863,
+      "step": 25
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 3.52e-05,
+      "loss": 0.7268,
+      "step": 30
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.6903731226921082,
+      "eval_runtime": 0.3504,
+      "eval_samples_per_second": 202.643,
+      "eval_steps_per_second": 2.854,
+      "step": 30
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 3.44e-05,
+      "loss": 0.7028,
+      "step": 35
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 0.7271296381950378,
+      "eval_runtime": 0.3505,
+      "eval_samples_per_second": 202.57,
+      "eval_steps_per_second": 2.853,
+      "step": 35
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 0.7096,
+      "step": 40
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5633802816901409,
+      "eval_loss": 0.6869621872901917,
+      "eval_runtime": 0.3521,
+      "eval_samples_per_second": 201.644,
+      "eval_steps_per_second": 2.84,
+      "step": 40
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 3.28e-05,
+      "loss": 0.6953,
+      "step": 45
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.43661971830985913,
+      "eval_loss": 0.7184516787528992,
+      "eval_runtime": 0.3498,
+      "eval_samples_per_second": 202.959,
+      "eval_steps_per_second": 2.859,
+      "step": 45
+    },
+    {
+      "epoch": 9.0,
+      "step": 45,
+      "total_flos": 847231498321920.0,
+      "train_loss": 0.7311807526482477,
+      "train_runtime": 109.9802,
+      "train_samples_per_second": 288.688,
+      "train_steps_per_second": 2.273
+    }
+  ],
+  "max_steps": 250,
+  "num_train_epochs": 50,
+  "total_flos": 847231498321920.0,
+  "trial_name": null,
+  "trial_params": null
+}