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: 0.8760079149146673
+      value: 0.8651001731387583
     - name: F1
       type: f1
-      value: 0.8322233006459385
+      value: 0.8160291438979962
 ---
 
 <!-- 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_qqp
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2](https://huggingface.co/gokuls/bert_12_layer_model_v2) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2](https://huggingface.co/gokuls/bert_12_layer_model_v2) on the GLUE QQP dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4723
-- Accuracy: 0.8760
-- F1: 0.8322
-- Combined Score: 0.8541
+- Loss: 0.3129
+- Accuracy: 0.8651
+- F1: 0.8160
+- Combined Score: 0.8406
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.8651001731387583,
+    "eval_combined_score": 0.8405646585183773,
+    "eval_f1": 0.8160291438979962,
+    "eval_loss": 0.31285223364830017,
+    "eval_runtime": 47.3472,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 853.905,
+    "eval_steps_per_second": 3.337,
+    "train_loss": 0.18701694115640075,
+    "train_runtime": 10256.0914,
+    "train_samples": 363846,
+    "train_samples_per_second": 1773.804,
+    "train_steps_per_second": 6.932
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.8651001731387583,
+    "eval_combined_score": 0.8405646585183773,
+    "eval_f1": 0.8160291438979962,
+    "eval_loss": 0.31285223364830017,
+    "eval_runtime": 47.3472,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 853.905,
+    "eval_steps_per_second": 3.337
+}

logs/events.out.tfevents.1677677003.kyoto.1072791.11

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:999d0a5027d71d44aa5bb2002cbac6f4dae9f2c5486353fdccdb661d661a7b83
+size 467

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 8.0,
+    "train_loss": 0.18701694115640075,
+    "train_runtime": 10256.0914,
+    "train_samples": 363846,
+    "train_samples_per_second": 1773.804,
+    "train_steps_per_second": 6.932
+}

trainer_state.json

@@ -0,0 +1,161 @@
+{
+  "best_metric": 0.31285223364830017,
+  "best_model_checkpoint": "hBERTv2_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": 4.900070323488045e-05,
+      "loss": 0.4179,
+      "step": 1422
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.8251545881770962,
+      "eval_combined_score": 0.8083731607119745,
+      "eval_f1": 0.7915917332468528,
+      "eval_loss": 0.38297852873802185,
+      "eval_runtime": 47.7223,
+      "eval_samples_per_second": 847.193,
+      "eval_steps_per_second": 3.311,
+      "step": 1422
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.800070323488045e-05,
+      "loss": 0.2978,
+      "step": 2844
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8357160524363096,
+      "eval_combined_score": 0.8131408666041042,
+      "eval_f1": 0.7905656807718988,
+      "eval_loss": 0.35074833035469055,
+      "eval_runtime": 47.7241,
+      "eval_samples_per_second": 847.161,
+      "eval_steps_per_second": 3.311,
+      "step": 2844
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.700070323488045e-05,
+      "loss": 0.2318,
+      "step": 4266
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.8651001731387583,
+      "eval_combined_score": 0.8405646585183773,
+      "eval_f1": 0.8160291438979962,
+      "eval_loss": 0.31285223364830017,
+      "eval_runtime": 47.7786,
+      "eval_samples_per_second": 846.194,
+      "eval_steps_per_second": 3.307,
+      "step": 4266
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.600070323488046e-05,
+      "loss": 0.1765,
+      "step": 5688
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8700469948058372,
+      "eval_combined_score": 0.8514036858419307,
+      "eval_f1": 0.8327603768780241,
+      "eval_loss": 0.35401350259780884,
+      "eval_runtime": 47.7873,
+      "eval_samples_per_second": 846.041,
+      "eval_steps_per_second": 3.306,
+      "step": 5688
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.500070323488045e-05,
+      "loss": 0.1305,
+      "step": 7110
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.8733860994311156,
+      "eval_combined_score": 0.8500331546757744,
+      "eval_f1": 0.8266802099204335,
+      "eval_loss": 0.4275738596916199,
+      "eval_runtime": 47.709,
+      "eval_samples_per_second": 847.43,
+      "eval_steps_per_second": 3.312,
+      "step": 7110
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.40014064697609e-05,
+      "loss": 0.1003,
+      "step": 8532
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.8748454118229038,
+      "eval_combined_score": 0.8520305093406406,
+      "eval_f1": 0.8292156068583774,
+      "eval_loss": 0.40782323479652405,
+      "eval_runtime": 47.8778,
+      "eval_samples_per_second": 844.441,
+      "eval_steps_per_second": 3.3,
+      "step": 8532
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.30014064697609e-05,
+      "loss": 0.0788,
+      "step": 9954
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.876675735839723,
+      "eval_combined_score": 0.855563447307603,
+      "eval_f1": 0.8344511587754831,
+      "eval_loss": 0.4068982005119324,
+      "eval_runtime": 47.8891,
+      "eval_samples_per_second": 844.242,
+      "eval_steps_per_second": 3.299,
+      "step": 9954
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.200210970464135e-05,
+      "loss": 0.0625,
+      "step": 11376
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.8760079149146673,
+      "eval_combined_score": 0.8541156077803029,
+      "eval_f1": 0.8322233006459385,
+      "eval_loss": 0.4722978472709656,
+      "eval_runtime": 47.7389,
+      "eval_samples_per_second": 846.898,
+      "eval_steps_per_second": 3.31,
+      "step": 11376
+    },
+    {
+      "epoch": 8.0,
+      "step": 11376,
+      "total_flos": 3.625461609146614e+17,
+      "train_loss": 0.18701694115640075,
+      "train_runtime": 10256.0914,
+      "train_samples_per_second": 1773.804,
+      "train_steps_per_second": 6.932
+    }
+  ],
+  "max_steps": 71100,
+  "num_train_epochs": 50,
+  "total_flos": 3.625461609146614e+17,
+  "trial_name": null,
+  "trial_params": null
+}