End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 tags:
 - generated_from_trainer
 datasets:
@@ -13,7 +15,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE QQP
       type: glue
       config: qqp
       split: validation
@@ -21,10 +23,10 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.8300766757358398
+      value: 0.8216176106851348
     - name: F1
       type: f1
-      value: 0.7659443990188063
+      value: 0.7561536380849337
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -32,12 +34,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hBERTv2_new_pretrain_48_qqp
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new_48](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new_48) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new_48](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new_48) on the GLUE QQP dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4908
-- Accuracy: 0.8301
-- F1: 0.7659
-- Combined Score: 0.7980
+- Loss: 0.4029
+- Accuracy: 0.8216
+- F1: 0.7562
+- Combined Score: 0.7889
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 9.0,
+    "eval_accuracy": 0.8216176106851348,
+    "eval_combined_score": 0.7888856243850343,
+    "eval_f1": 0.7561536380849337,
+    "eval_loss": 0.40290606021881104,
+    "eval_runtime": 66.1981,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 610.743,
+    "eval_steps_per_second": 4.774,
+    "train_loss": 0.3016081741529207,
+    "train_runtime": 15759.4372,
+    "train_samples": 363846,
+    "train_samples_per_second": 1154.375,
+    "train_steps_per_second": 9.02
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 9.0,
+    "eval_accuracy": 0.8216176106851348,
+    "eval_combined_score": 0.7888856243850343,
+    "eval_f1": 0.7561536380849337,
+    "eval_loss": 0.40290606021881104,
+    "eval_runtime": 66.1981,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 610.743,
+    "eval_steps_per_second": 4.774
+}

logs/events.out.tfevents.1686053614.serv-9207.2387813.14

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d93de99dd03b32918dcc969b92585ba89842bc64c2a73bd42c163836b9c3dad8
+size 475

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 9.0,
+    "train_loss": 0.3016081741529207,
+    "train_runtime": 15759.4372,
+    "train_samples": 363846,
+    "train_samples_per_second": 1154.375,
+    "train_steps_per_second": 9.02
+}

trainer_state.json

@@ -0,0 +1,178 @@
+{
+  "best_metric": 0.40290606021881104,
+  "best_model_checkpoint": "hBERTv2_new_pretrain_48_qqp/checkpoint-11372",
+  "epoch": 9.0,
+  "global_step": 25587,
+  "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": 0.5044,
+      "step": 2843
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.78651991095721,
+      "eval_combined_score": 0.7413003794371951,
+      "eval_f1": 0.6960808479171803,
+      "eval_loss": 0.44682765007019043,
+      "eval_runtime": 66.1757,
+      "eval_samples_per_second": 610.949,
+      "eval_steps_per_second": 4.775,
+      "step": 2843
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.4102,
+      "step": 5686
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.7992332426416028,
+      "eval_combined_score": 0.746378939614175,
+      "eval_f1": 0.6935246365867472,
+      "eval_loss": 0.43585720658302307,
+      "eval_runtime": 65.9585,
+      "eval_samples_per_second": 612.961,
+      "eval_steps_per_second": 4.791,
+      "step": 5686
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.76e-05,
+      "loss": 0.3553,
+      "step": 8529
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.8080138511006678,
+      "eval_combined_score": 0.7592367077049147,
+      "eval_f1": 0.7104595643091615,
+      "eval_loss": 0.41269394755363464,
+      "eval_runtime": 66.1428,
+      "eval_samples_per_second": 611.253,
+      "eval_steps_per_second": 4.778,
+      "step": 8529
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.3122,
+      "step": 11372
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8216176106851348,
+      "eval_combined_score": 0.7888856243850343,
+      "eval_f1": 0.7561536380849337,
+      "eval_loss": 0.40290606021881104,
+      "eval_runtime": 67.1707,
+      "eval_samples_per_second": 601.899,
+      "eval_steps_per_second": 4.704,
+      "step": 11372
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.2756,
+      "step": 14215
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.8227801137768983,
+      "eval_combined_score": 0.7873036033537604,
+      "eval_f1": 0.7518270929306224,
+      "eval_loss": 0.4481242597103119,
+      "eval_runtime": 66.1409,
+      "eval_samples_per_second": 611.271,
+      "eval_steps_per_second": 4.778,
+      "step": 14215
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 3.52e-05,
+      "loss": 0.2479,
+      "step": 17058
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.8267870393272323,
+      "eval_combined_score": 0.795055623823946,
+      "eval_f1": 0.7633242083206597,
+      "eval_loss": 0.47777727246284485,
+      "eval_runtime": 68.5855,
+      "eval_samples_per_second": 589.483,
+      "eval_steps_per_second": 4.607,
+      "step": 17058
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 3.44e-05,
+      "loss": 0.223,
+      "step": 19901
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.8157803611179817,
+      "eval_combined_score": 0.793922742684098,
+      "eval_f1": 0.7720651242502142,
+      "eval_loss": 0.44251149892807007,
+      "eval_runtime": 66.188,
+      "eval_samples_per_second": 610.835,
+      "eval_steps_per_second": 4.774,
+      "step": 19901
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 0.2028,
+      "step": 22744
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.8266881028938907,
+      "eval_combined_score": 0.7976520790531931,
+      "eval_f1": 0.7686160552124955,
+      "eval_loss": 0.4705200493335724,
+      "eval_runtime": 67.2811,
+      "eval_samples_per_second": 600.912,
+      "eval_steps_per_second": 4.697,
+      "step": 22744
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 3.28e-05,
+      "loss": 0.183,
+      "step": 25587
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.8300766757358398,
+      "eval_combined_score": 0.798010537377323,
+      "eval_f1": 0.7659443990188063,
+      "eval_loss": 0.4907599687576294,
+      "eval_runtime": 67.1619,
+      "eval_samples_per_second": 601.979,
+      "eval_steps_per_second": 4.705,
+      "step": 25587
+    },
+    {
+      "epoch": 9.0,
+      "step": 25587,
+      "total_flos": 4.7920428019718554e+17,
+      "train_loss": 0.3016081741529207,
+      "train_runtime": 15759.4372,
+      "train_samples_per_second": 1154.375,
+      "train_steps_per_second": 9.02
+    }
+  ],
+  "max_steps": 142150,
+  "num_train_epochs": 50,
+  "total_flos": 4.7920428019718554e+17,
+  "trial_name": null,
+  "trial_params": null
+}