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: 0.8781103141231759
+      value: 0.8679940638139995
     - name: F1
       type: f1
-      value: 0.8354371201496026
+      value: 0.8221652060910999
 ---
 
 <!-- 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. -->
 
 # hBERTv1_qqp
 
-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 QQP dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4176
-- Accuracy: 0.8781
-- F1: 0.8354
-- Combined Score: 0.8568
+- Loss: 0.3039
+- Accuracy: 0.8680
+- F1: 0.8222
+- Combined Score: 0.8451
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.8679940638139995,
+    "eval_combined_score": 0.8450796349525497,
+    "eval_f1": 0.8221652060910999,
+    "eval_loss": 0.3039037883281708,
+    "eval_runtime": 49.9948,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 808.685,
+    "eval_steps_per_second": 3.16,
+    "train_loss": 0.20186867962071328,
+    "train_runtime": 10978.3186,
+    "train_samples": 363846,
+    "train_samples_per_second": 1657.112,
+    "train_steps_per_second": 6.476
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.8679940638139995,
+    "eval_combined_score": 0.8450796349525497,
+    "eval_f1": 0.8221652060910999,
+    "eval_loss": 0.3039037883281708,
+    "eval_runtime": 49.9948,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 808.685,
+    "eval_steps_per_second": 3.16
+}

logs/events.out.tfevents.1677696213.serv-9220.1178704.11

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d26718a84b02fc78aba35a56f4dc2e2ddd0d0fcf4f1a5e7735e39b6c206ed2b
+size 467

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 8.0,
+    "train_loss": 0.20186867962071328,
+    "train_runtime": 10978.3186,
+    "train_samples": 363846,
+    "train_samples_per_second": 1657.112,
+    "train_steps_per_second": 6.476
+}

trainer_state.json

@@ -0,0 +1,161 @@
+{
+  "best_metric": 0.3039037883281708,
+  "best_model_checkpoint": "hBERTv1_qqp/checkpoint-4266",
+  "epoch": 8.0,
+  "global_step": 11376,
+  "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": 0.4011,
+      "step": 1422
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.8285678951273806,
+      "eval_combined_score": 0.8116115925137688,
+      "eval_f1": 0.794655289900157,
+      "eval_loss": 0.3665443956851959,
+      "eval_runtime": 50.6077,
+      "eval_samples_per_second": 798.89,
+      "eval_steps_per_second": 3.122,
+      "step": 1422
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.8e-05,
+      "loss": 0.3026,
+      "step": 2844
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8625278258718774,
+      "eval_combined_score": 0.8398193588062908,
+      "eval_f1": 0.8171108917407043,
+      "eval_loss": 0.3111252784729004,
+      "eval_runtime": 50.4355,
+      "eval_samples_per_second": 801.618,
+      "eval_steps_per_second": 3.133,
+      "step": 2844
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7e-05,
+      "loss": 0.2472,
+      "step": 4266
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.8679940638139995,
+      "eval_combined_score": 0.8450796349525497,
+      "eval_f1": 0.8221652060910999,
+      "eval_loss": 0.3039037883281708,
+      "eval_runtime": 50.4592,
+      "eval_samples_per_second": 801.241,
+      "eval_steps_per_second": 3.131,
+      "step": 4266
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.1983,
+      "step": 5688
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8736581746228048,
+      "eval_combined_score": 0.8531681858438971,
+      "eval_f1": 0.8326781970649895,
+      "eval_loss": 0.32322049140930176,
+      "eval_runtime": 50.5899,
+      "eval_samples_per_second": 799.171,
+      "eval_steps_per_second": 3.123,
+      "step": 5688
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.500070323488045e-05,
+      "loss": 0.157,
+      "step": 7110
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.8717289141726441,
+      "eval_combined_score": 0.8455726540359703,
+      "eval_f1": 0.8194163938992965,
+      "eval_loss": 0.3742031753063202,
+      "eval_runtime": 50.4756,
+      "eval_samples_per_second": 800.98,
+      "eval_steps_per_second": 3.13,
+      "step": 7110
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.400070323488045e-05,
+      "loss": 0.1251,
+      "step": 8532
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.8716299777393025,
+      "eval_combined_score": 0.8431165534906415,
+      "eval_f1": 0.8146031292419805,
+      "eval_loss": 0.40085598826408386,
+      "eval_runtime": 50.493,
+      "eval_samples_per_second": 800.705,
+      "eval_steps_per_second": 3.129,
+      "step": 8532
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.30014064697609e-05,
+      "loss": 0.1009,
+      "step": 9954
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.8699480583724957,
+      "eval_combined_score": 0.8499536553423827,
+      "eval_f1": 0.8299592523122696,
+      "eval_loss": 0.4470857083797455,
+      "eval_runtime": 50.5119,
+      "eval_samples_per_second": 800.405,
+      "eval_steps_per_second": 3.128,
+      "step": 9954
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.200210970464135e-05,
+      "loss": 0.0828,
+      "step": 11376
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.8781103141231759,
+      "eval_combined_score": 0.8567737171363892,
+      "eval_f1": 0.8354371201496026,
+      "eval_loss": 0.4176293909549713,
+      "eval_runtime": 50.4947,
+      "eval_samples_per_second": 800.677,
+      "eval_steps_per_second": 3.129,
+      "step": 11376
+    },
+    {
+      "epoch": 8.0,
+      "step": 11376,
+      "total_flos": 3.676621197992264e+17,
+      "train_loss": 0.20186867962071328,
+      "train_runtime": 10978.3186,
+      "train_samples_per_second": 1657.112,
+      "train_steps_per_second": 6.476
+    }
+  ],
+  "max_steps": 71100,
+  "num_train_epochs": 50,
+  "total_flos": 3.676621197992264e+17,
+  "trial_name": null,
+  "trial_params": null
+}