End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 base_model: gokuls/bert_12_layer_model_v2_complete_training_new_48
 tags:
 - generated_from_trainer
@@ -13,7 +15,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE STSB
       type: glue
       config: stsb
       split: validation
@@ -21,7 +23,7 @@ model-index:
     metrics:
     - name: Spearmanr
       type: spearmanr
-      value: 0.3808984066515933
+      value: 0.38294286179733106
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -29,12 +31,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hBERTv2_new_pretrain_48_ver2_stsb
 
-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 STSB dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.4339
-- Pearson: 0.3873
-- Spearmanr: 0.3809
-- Combined Score: 0.3841
+- Loss: 2.0659
+- Pearson: 0.3927
+- Spearmanr: 0.3829
+- Combined Score: 0.3878
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 12.0,
+    "eval_combined_score": 0.3878159990767361,
+    "eval_loss": 2.065941333770752,
+    "eval_pearson": 0.3926891363561411,
+    "eval_runtime": 7.6775,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 195.375,
+    "eval_spearmanr": 0.38294286179733106,
+    "eval_steps_per_second": 3.126,
+    "train_loss": 1.043103571291323,
+    "train_runtime": 1213.9732,
+    "train_samples": 5749,
+    "train_samples_per_second": 71.035,
+    "train_steps_per_second": 1.112
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 12.0,
+    "eval_combined_score": 0.3878159990767361,
+    "eval_loss": 2.065941333770752,
+    "eval_pearson": 0.3926891363561411,
+    "eval_runtime": 7.6775,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 195.375,
+    "eval_spearmanr": 0.38294286179733106,
+    "eval_steps_per_second": 3.126
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 12.0,
+    "train_loss": 1.043103571291323,
+    "train_runtime": 1213.9732,
+    "train_samples": 5749,
+    "train_samples_per_second": 71.035,
+    "train_steps_per_second": 1.112
+}

trainer_state.json

@@ -0,0 +1,232 @@
+{
+  "best_metric": 2.065941333770752,
+  "best_model_checkpoint": "hBERTv2_new_pretrain_48_ver2_stsb/checkpoint-630",
+  "epoch": 12.0,
+  "eval_steps": 500,
+  "global_step": 1080,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.733333333333334e-05,
+      "loss": 2.3158,
+      "step": 90
+    },
+    {
+      "epoch": 1.0,
+      "eval_combined_score": 0.12682569890318168,
+      "eval_loss": 2.778519630432129,
+      "eval_pearson": 0.13553557974356426,
+      "eval_runtime": 7.873,
+      "eval_samples_per_second": 190.525,
+      "eval_spearmanr": 0.11811581806279914,
+      "eval_steps_per_second": 3.048,
+      "step": 90
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.466666666666667e-05,
+      "loss": 2.0193,
+      "step": 180
+    },
+    {
+      "epoch": 2.0,
+      "eval_combined_score": 0.20018816275647738,
+      "eval_loss": 2.656583547592163,
+      "eval_pearson": 0.1953722464972577,
+      "eval_runtime": 7.8712,
+      "eval_samples_per_second": 190.568,
+      "eval_spearmanr": 0.20500407901569703,
+      "eval_steps_per_second": 3.049,
+      "step": 180
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 1.7471,
+      "step": 270
+    },
+    {
+      "epoch": 3.0,
+      "eval_combined_score": 0.26123838219196927,
+      "eval_loss": 2.5201523303985596,
+      "eval_pearson": 0.25777290222896787,
+      "eval_runtime": 7.8895,
+      "eval_samples_per_second": 190.127,
+      "eval_spearmanr": 0.26470386215497066,
+      "eval_steps_per_second": 3.042,
+      "step": 270
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 2.9333333333333333e-05,
+      "loss": 1.5032,
+      "step": 360
+    },
+    {
+      "epoch": 4.0,
+      "eval_combined_score": 0.2668439145531881,
+      "eval_loss": 3.060143232345581,
+      "eval_pearson": 0.2604800073536366,
+      "eval_runtime": 7.903,
+      "eval_samples_per_second": 189.801,
+      "eval_spearmanr": 0.27320782175273967,
+      "eval_steps_per_second": 3.037,
+      "step": 360
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 2.6666666666666667e-05,
+      "loss": 1.1825,
+      "step": 450
+    },
+    {
+      "epoch": 5.0,
+      "eval_combined_score": 0.3165039136131461,
+      "eval_loss": 2.337822675704956,
+      "eval_pearson": 0.31504430380844267,
+      "eval_runtime": 7.883,
+      "eval_samples_per_second": 190.284,
+      "eval_spearmanr": 0.3179635234178495,
+      "eval_steps_per_second": 3.045,
+      "step": 450
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 2.4e-05,
+      "loss": 0.8788,
+      "step": 540
+    },
+    {
+      "epoch": 6.0,
+      "eval_combined_score": 0.3429143690890755,
+      "eval_loss": 2.365739583969116,
+      "eval_pearson": 0.3437239899365901,
+      "eval_runtime": 7.8845,
+      "eval_samples_per_second": 190.247,
+      "eval_spearmanr": 0.34210474824156095,
+      "eval_steps_per_second": 3.044,
+      "step": 540
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 2.1333333333333335e-05,
+      "loss": 0.6987,
+      "step": 630
+    },
+    {
+      "epoch": 7.0,
+      "eval_combined_score": 0.3878159990767361,
+      "eval_loss": 2.065941333770752,
+      "eval_pearson": 0.3926891363561411,
+      "eval_runtime": 7.868,
+      "eval_samples_per_second": 190.647,
+      "eval_spearmanr": 0.38294286179733106,
+      "eval_steps_per_second": 3.05,
+      "step": 630
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 1.866666666666667e-05,
+      "loss": 0.5879,
+      "step": 720
+    },
+    {
+      "epoch": 8.0,
+      "eval_combined_score": 0.36337847108214494,
+      "eval_loss": 2.671151638031006,
+      "eval_pearson": 0.3631367983109315,
+      "eval_runtime": 7.8716,
+      "eval_samples_per_second": 190.559,
+      "eval_spearmanr": 0.36362014385335845,
+      "eval_steps_per_second": 3.049,
+      "step": 720
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 1.6000000000000003e-05,
+      "loss": 0.4865,
+      "step": 810
+    },
+    {
+      "epoch": 9.0,
+      "eval_combined_score": 0.3645136865267287,
+      "eval_loss": 2.306649684906006,
+      "eval_pearson": 0.36653331894419605,
+      "eval_runtime": 7.88,
+      "eval_samples_per_second": 190.356,
+      "eval_spearmanr": 0.3624940541092614,
+      "eval_steps_per_second": 3.046,
+      "step": 810
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 1.3333333333333333e-05,
+      "loss": 0.4233,
+      "step": 900
+    },
+    {
+      "epoch": 10.0,
+      "eval_combined_score": 0.37241863884244814,
+      "eval_loss": 2.2780683040618896,
+      "eval_pearson": 0.37533403902944973,
+      "eval_runtime": 7.8584,
+      "eval_samples_per_second": 190.879,
+      "eval_spearmanr": 0.36950323865544654,
+      "eval_steps_per_second": 3.054,
+      "step": 900
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 1.0666666666666667e-05,
+      "loss": 0.3628,
+      "step": 990
+    },
+    {
+      "epoch": 11.0,
+      "eval_combined_score": 0.38028607037665435,
+      "eval_loss": 2.467188835144043,
+      "eval_pearson": 0.38475327674855436,
+      "eval_runtime": 7.861,
+      "eval_samples_per_second": 190.814,
+      "eval_spearmanr": 0.37581886400475434,
+      "eval_steps_per_second": 3.053,
+      "step": 990
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 0.3113,
+      "step": 1080
+    },
+    {
+      "epoch": 12.0,
+      "eval_combined_score": 0.38409591427031303,
+      "eval_loss": 2.433856964111328,
+      "eval_pearson": 0.38729342188903276,
+      "eval_runtime": 7.854,
+      "eval_samples_per_second": 190.987,
+      "eval_spearmanr": 0.3808984066515933,
+      "eval_steps_per_second": 3.056,
+      "step": 1080
+    },
+    {
+      "epoch": 12.0,
+      "step": 1080,
+      "total_flos": 1.0095565442383872e+16,
+      "train_loss": 1.043103571291323,
+      "train_runtime": 1213.9732,
+      "train_samples_per_second": 71.035,
+      "train_steps_per_second": 1.112
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 1350,
+  "num_train_epochs": 15,
+  "save_steps": 500,
+  "total_flos": 1.0095565442383872e+16,
+  "trial_name": null,
+  "trial_params": null
+}