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: 0.8112787534009399
+      value: 0.8048973534504081
     - name: F1
       type: f1
-      value: 0.7559961624560281
+      value: 0.7301771909420538
 ---
 
 <!-- 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. -->
 
 # add_BERT_24_qqp
 
-This model is a fine-tuned version of [gokuls/add_bert_12_layer_model_complete_training_new](https://huggingface.co/gokuls/add_bert_12_layer_model_complete_training_new) on the glue dataset.
+This model is a fine-tuned version of [gokuls/add_bert_12_layer_model_complete_training_new](https://huggingface.co/gokuls/add_bert_12_layer_model_complete_training_new) on the GLUE QQP dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5186
-- Accuracy: 0.8113
-- F1: 0.7560
-- Combined Score: 0.7836
+- Loss: 0.4356
+- Accuracy: 0.8049
+- F1: 0.7302
+- Combined Score: 0.7675
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 18.0,
+    "eval_accuracy": 0.8048973534504081,
+    "eval_combined_score": 0.767537272196231,
+    "eval_f1": 0.7301771909420538,
+    "eval_loss": 0.4356277883052826,
+    "eval_runtime": 76.3096,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 529.815,
+    "eval_steps_per_second": 4.141,
+    "train_loss": 0.38470190150118955,
+    "train_runtime": 39986.3174,
+    "train_samples": 363846,
+    "train_samples_per_second": 454.963,
+    "train_steps_per_second": 3.555
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 18.0,
+    "eval_accuracy": 0.8048973534504081,
+    "eval_combined_score": 0.767537272196231,
+    "eval_f1": 0.7301771909420538,
+    "eval_loss": 0.4356277883052826,
+    "eval_runtime": 76.3096,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 529.815,
+    "eval_steps_per_second": 4.141
+}

logs/events.out.tfevents.1686913379.serv-3317.1852327.7

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9209b80a7c9681543acb6487219eee351b0254ce436aea14d8e01f60c397fe88
+size 475

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 18.0,
+    "train_loss": 0.38470190150118955,
+    "train_runtime": 39986.3174,
+    "train_samples": 363846,
+    "train_samples_per_second": 454.963,
+    "train_steps_per_second": 3.555
+}

trainer_state.json

@@ -0,0 +1,331 @@
+{
+  "best_metric": 0.4356277883052826,
+  "best_model_checkpoint": "add_BERT_24_qqp/checkpoint-36959",
+  "epoch": 18.0,
+  "global_step": 51174,
+  "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": 0.5487,
+      "step": 2843
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7476873608706406,
+      "eval_combined_score": 0.6971173022779453,
+      "eval_f1": 0.64654724368525,
+      "eval_loss": 0.5164108276367188,
+      "eval_runtime": 76.9084,
+      "eval_samples_per_second": 525.69,
+      "eval_steps_per_second": 4.109,
+      "step": 2843
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.4981,
+      "step": 5686
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.7635419243136284,
+      "eval_combined_score": 0.7061260734965658,
+      "eval_f1": 0.6487102226795032,
+      "eval_loss": 0.4939361810684204,
+      "eval_runtime": 76.8599,
+      "eval_samples_per_second": 526.022,
+      "eval_steps_per_second": 4.111,
+      "step": 5686
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.76e-05,
+      "loss": 0.4835,
+      "step": 8529
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.7568142468464012,
+      "eval_combined_score": 0.6855623093638878,
+      "eval_f1": 0.6143103718813745,
+      "eval_loss": 0.4989795386791229,
+      "eval_runtime": 76.8916,
+      "eval_samples_per_second": 525.805,
+      "eval_steps_per_second": 4.11,
+      "step": 8529
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.4719,
+      "step": 11372
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7636655948553055,
+      "eval_combined_score": 0.7026922222279828,
+      "eval_f1": 0.64171884960066,
+      "eval_loss": 0.491183876991272,
+      "eval_runtime": 76.8632,
+      "eval_samples_per_second": 525.999,
+      "eval_steps_per_second": 4.111,
+      "step": 11372
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.4632,
+      "step": 14215
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.7680435320306703,
+      "eval_combined_score": 0.7149878799302595,
+      "eval_f1": 0.6619322278298486,
+      "eval_loss": 0.48811477422714233,
+      "eval_runtime": 76.8825,
+      "eval_samples_per_second": 525.867,
+      "eval_steps_per_second": 4.11,
+      "step": 14215
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 3.52e-05,
+      "loss": 0.4584,
+      "step": 17058
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.7679198614889933,
+      "eval_combined_score": 0.7129432719758604,
+      "eval_f1": 0.6579666824627274,
+      "eval_loss": 0.483871728181839,
+      "eval_runtime": 76.9228,
+      "eval_samples_per_second": 525.592,
+      "eval_steps_per_second": 4.108,
+      "step": 17058
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 3.44e-05,
+      "loss": 0.4425,
+      "step": 19901
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.7723225327726936,
+      "eval_combined_score": 0.7318651116376123,
+      "eval_f1": 0.691407690502531,
+      "eval_loss": 0.47737881541252136,
+      "eval_runtime": 76.8111,
+      "eval_samples_per_second": 526.356,
+      "eval_steps_per_second": 4.114,
+      "step": 19901
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 0.4308,
+      "step": 22744
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.7738065792728173,
+      "eval_combined_score": 0.7194063117251298,
+      "eval_f1": 0.6650060441774424,
+      "eval_loss": 0.4679051637649536,
+      "eval_runtime": 76.7813,
+      "eval_samples_per_second": 526.561,
+      "eval_steps_per_second": 4.116,
+      "step": 22744
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 3.28e-05,
+      "loss": 0.4102,
+      "step": 25587
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.7872866683156072,
+      "eval_combined_score": 0.7393444967816015,
+      "eval_f1": 0.6914023252475958,
+      "eval_loss": 0.45359379053115845,
+      "eval_runtime": 76.7919,
+      "eval_samples_per_second": 526.488,
+      "eval_steps_per_second": 4.115,
+      "step": 25587
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.3909,
+      "step": 28430
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.7895127380657927,
+      "eval_combined_score": 0.7524201142660535,
+      "eval_f1": 0.7153274904663143,
+      "eval_loss": 0.4511614739894867,
+      "eval_runtime": 76.7568,
+      "eval_samples_per_second": 526.729,
+      "eval_steps_per_second": 4.117,
+      "step": 28430
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 3.1200000000000006e-05,
+      "loss": 0.3787,
+      "step": 31273
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.7959436062329953,
+      "eval_combined_score": 0.7546729772983917,
+      "eval_f1": 0.713402348363788,
+      "eval_loss": 0.46807020902633667,
+      "eval_runtime": 76.8356,
+      "eval_samples_per_second": 526.188,
+      "eval_steps_per_second": 4.113,
+      "step": 31273
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 3.0400000000000004e-05,
+      "loss": 0.3538,
+      "step": 34116
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.7980707395498392,
+      "eval_combined_score": 0.7537684658603288,
+      "eval_f1": 0.7094661921708184,
+      "eval_loss": 0.44865256547927856,
+      "eval_runtime": 76.8015,
+      "eval_samples_per_second": 526.422,
+      "eval_steps_per_second": 4.115,
+      "step": 34116
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 2.96e-05,
+      "loss": 0.3313,
+      "step": 36959
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.8048973534504081,
+      "eval_combined_score": 0.767537272196231,
+      "eval_f1": 0.7301771909420538,
+      "eval_loss": 0.4356277883052826,
+      "eval_runtime": 76.7753,
+      "eval_samples_per_second": 526.602,
+      "eval_steps_per_second": 4.116,
+      "step": 36959
+    },
+    {
+      "epoch": 14.0,
+      "learning_rate": 2.8800000000000002e-05,
+      "loss": 0.3053,
+      "step": 39802
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.8080633193173387,
+      "eval_combined_score": 0.7764252661836972,
+      "eval_f1": 0.7447872130500558,
+      "eval_loss": 0.44097900390625,
+      "eval_runtime": 77.018,
+      "eval_samples_per_second": 524.942,
+      "eval_steps_per_second": 4.103,
+      "step": 39802
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 2.8e-05,
+      "loss": 0.2785,
+      "step": 42645
+    },
+    {
+      "epoch": 15.0,
+      "eval_accuracy": 0.7942122186495176,
+      "eval_combined_score": 0.7696160694841214,
+      "eval_f1": 0.7450199203187251,
+      "eval_loss": 0.489566445350647,
+      "eval_runtime": 76.9319,
+      "eval_samples_per_second": 525.53,
+      "eval_steps_per_second": 4.108,
+      "step": 42645
+    },
+    {
+      "epoch": 16.0,
+      "learning_rate": 2.7200000000000004e-05,
+      "loss": 0.2516,
+      "step": 45488
+    },
+    {
+      "epoch": 16.0,
+      "eval_accuracy": 0.8054909720504576,
+      "eval_combined_score": 0.7782284171798302,
+      "eval_f1": 0.7509658623092027,
+      "eval_loss": 0.49687308073043823,
+      "eval_runtime": 76.9432,
+      "eval_samples_per_second": 525.452,
+      "eval_steps_per_second": 4.107,
+      "step": 45488
+    },
+    {
+      "epoch": 17.0,
+      "learning_rate": 2.6400000000000005e-05,
+      "loss": 0.2254,
+      "step": 48331
+    },
+    {
+      "epoch": 17.0,
+      "eval_accuracy": 0.8128864704427405,
+      "eval_combined_score": 0.7831825710964966,
+      "eval_f1": 0.7534786717502525,
+      "eval_loss": 0.507929801940918,
+      "eval_runtime": 76.8795,
+      "eval_samples_per_second": 525.888,
+      "eval_steps_per_second": 4.11,
+      "step": 48331
+    },
+    {
+      "epoch": 18.0,
+      "learning_rate": 2.5600000000000002e-05,
+      "loss": 0.2017,
+      "step": 51174
+    },
+    {
+      "epoch": 18.0,
+      "eval_accuracy": 0.8112787534009399,
+      "eval_combined_score": 0.7836374579284839,
+      "eval_f1": 0.7559961624560281,
+      "eval_loss": 0.5186213850975037,
+      "eval_runtime": 76.7033,
+      "eval_samples_per_second": 527.096,
+      "eval_steps_per_second": 4.12,
+      "step": 51174
+    },
+    {
+      "epoch": 18.0,
+      "step": 51174,
+      "total_flos": 9.353867454138286e+17,
+      "train_loss": 0.38470190150118955,
+      "train_runtime": 39986.3174,
+      "train_samples_per_second": 454.963,
+      "train_steps_per_second": 3.555
+    }
+  ],
+  "max_steps": 142150,
+  "num_train_epochs": 50,
+  "total_flos": 9.353867454138286e+17,
+  "trial_name": null,
+  "trial_params": null
+}