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 QNLI
       type: glue
       config: qnli
       split: validation
@@ -20,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.5793520043931906
+      value: 0.5837451949478308
 ---
 
 <!-- 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_qnli
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_48](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_48) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_48](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_48) on the GLUE QNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7672
-- Accuracy: 0.5794
+- Loss: 0.6678
+- Accuracy: 0.5837
 
 ## Model description
 

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.5837451949478308,
+    "eval_loss": 0.6678495407104492,
+    "eval_runtime": 9.6387,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 566.778,
+    "eval_steps_per_second": 4.461,
+    "train_loss": 0.6315902105784221,
+    "train_runtime": 3878.247,
+    "train_samples": 104743,
+    "train_samples_per_second": 1350.391,
+    "train_steps_per_second": 10.559
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.5837451949478308,
+    "eval_loss": 0.6678495407104492,
+    "eval_runtime": 9.6387,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 566.778,
+    "eval_steps_per_second": 4.461
+}

logs/events.out.tfevents.1686038276.serv-9219.1456809.11

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da606fc72bd500d08c5a8c464935ae32f21c7a479a373a7e73386aec8472dcf6
+size 363

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 7.0,
+    "train_loss": 0.6315902105784221,
+    "train_runtime": 3878.247,
+    "train_samples": 104743,
+    "train_samples_per_second": 1350.391,
+    "train_steps_per_second": 10.559
+}

trainer_state.json

@@ -0,0 +1,130 @@
+{
+  "best_metric": 0.6678495407104492,
+  "best_model_checkpoint": "hBERTv1_new_pretrain_48_qnli/checkpoint-1638",
+  "epoch": 7.0,
+  "global_step": 5733,
+  "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": 0.6818,
+      "step": 819
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5815485996705108,
+      "eval_loss": 0.6781542897224426,
+      "eval_runtime": 9.6934,
+      "eval_samples_per_second": 563.577,
+      "eval_steps_per_second": 4.436,
+      "step": 819
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.6686,
+      "step": 1638
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5837451949478308,
+      "eval_loss": 0.6678495407104492,
+      "eval_runtime": 9.7146,
+      "eval_samples_per_second": 562.347,
+      "eval_steps_per_second": 4.426,
+      "step": 1638
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.76e-05,
+      "loss": 0.6472,
+      "step": 2457
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5846604429800476,
+      "eval_loss": 0.6737925410270691,
+      "eval_runtime": 9.726,
+      "eval_samples_per_second": 561.693,
+      "eval_steps_per_second": 4.421,
+      "step": 2457
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.6311,
+      "step": 3276
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5802672524254073,
+      "eval_loss": 0.6779181361198425,
+      "eval_runtime": 9.6997,
+      "eval_samples_per_second": 563.212,
+      "eval_steps_per_second": 4.433,
+      "step": 3276
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.6142,
+      "step": 4095
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5850265421929343,
+      "eval_loss": 0.6802344918251038,
+      "eval_runtime": 9.6984,
+      "eval_samples_per_second": 563.289,
+      "eval_steps_per_second": 4.434,
+      "step": 4095
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 3.52e-05,
+      "loss": 0.5969,
+      "step": 4914
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5861248398315944,
+      "eval_loss": 0.7075582146644592,
+      "eval_runtime": 9.6998,
+      "eval_samples_per_second": 563.206,
+      "eval_steps_per_second": 4.433,
+      "step": 4914
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 3.44e-05,
+      "loss": 0.5814,
+      "step": 5733
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5793520043931906,
+      "eval_loss": 0.7671993374824524,
+      "eval_runtime": 9.7166,
+      "eval_samples_per_second": 562.234,
+      "eval_steps_per_second": 4.425,
+      "step": 5733
+    },
+    {
+      "epoch": 7.0,
+      "step": 5733,
+      "total_flos": 1.085847006787666e+17,
+      "train_loss": 0.6315902105784221,
+      "train_runtime": 3878.247,
+      "train_samples_per_second": 1350.391,
+      "train_steps_per_second": 10.559
+    }
+  ],
+  "max_steps": 40950,
+  "num_train_epochs": 50,
+  "total_flos": 1.085847006787666e+17,
+  "trial_name": null,
+  "trial_params": null
+}