End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 tags:
 - generated_from_trainer
 datasets:
@@ -12,7 +14,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE STSB
       type: glue
       config: stsb
       split: validation
@@ -20,7 +22,7 @@ model-index:
     metrics:
     - name: Spearmanr
       type: spearmanr
-      value: 0.750517182024731
+      value: 0.7471924680940966
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -28,12 +30,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hBERTv1_new_pretrain_w_init_48_stsb
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_48](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_48) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_48](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_48) on the GLUE STSB dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0470
-- Pearson: 0.7517
-- Spearmanr: 0.7505
-- Combined Score: 0.7511
+- Loss: 0.9800
+- Pearson: 0.7515
+- Spearmanr: 0.7472
+- Combined Score: 0.7493
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 14.0,
+    "eval_combined_score": 0.7493306108739174,
+    "eval_loss": 0.9799954295158386,
+    "eval_pearson": 0.7514687536537382,
+    "eval_runtime": 2.8106,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 533.694,
+    "eval_spearmanr": 0.7471924680940966,
+    "eval_steps_per_second": 4.27,
+    "train_loss": 0.7678726120600625,
+    "train_runtime": 566.4946,
+    "train_samples": 5749,
+    "train_samples_per_second": 507.419,
+    "train_steps_per_second": 3.972
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 14.0,
+    "eval_combined_score": 0.7493306108739174,
+    "eval_loss": 0.9799954295158386,
+    "eval_pearson": 0.7514687536537382,
+    "eval_runtime": 2.8106,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 533.694,
+    "eval_spearmanr": 0.7471924680940966,
+    "eval_steps_per_second": 4.27
+}

logs/events.out.tfevents.1686058598.serv-9207.2389173.20

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5654bfc0dab8975c9dbbeb97a34160295c8a55cdb56f66a4bd5a60b24bedc09c
+size 473

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 14.0,
+    "train_loss": 0.7678726120600625,
+    "train_runtime": 566.4946,
+    "train_samples": 5749,
+    "train_samples_per_second": 507.419,
+    "train_steps_per_second": 3.972
+}

trainer_state.json

@@ -0,0 +1,263 @@
+{
+  "best_metric": 0.9799954295158386,
+  "best_model_checkpoint": "hBERTv1_new_pretrain_w_init_48_stsb/checkpoint-405",
+  "epoch": 14.0,
+  "global_step": 630,
+  "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": 2.5456,
+      "step": 45
+    },
+    {
+      "epoch": 1.0,
+      "eval_combined_score": 0.11935926461991596,
+      "eval_loss": 2.270618438720703,
+      "eval_pearson": 0.12461202621384566,
+      "eval_runtime": 2.794,
+      "eval_samples_per_second": 536.862,
+      "eval_spearmanr": 0.11410650302598625,
+      "eval_steps_per_second": 4.295,
+      "step": 45
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 2.0514,
+      "step": 90
+    },
+    {
+      "epoch": 2.0,
+      "eval_combined_score": 0.5231834246957023,
+      "eval_loss": 2.0612900257110596,
+      "eval_pearson": 0.5265750419588554,
+      "eval_runtime": 2.8043,
+      "eval_samples_per_second": 534.884,
+      "eval_spearmanr": 0.5197918074325493,
+      "eval_steps_per_second": 4.279,
+      "step": 90
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.76e-05,
+      "loss": 1.3837,
+      "step": 135
+    },
+    {
+      "epoch": 3.0,
+      "eval_combined_score": 0.6897094834383319,
+      "eval_loss": 1.1984349489212036,
+      "eval_pearson": 0.685263316928048,
+      "eval_runtime": 2.8304,
+      "eval_samples_per_second": 529.967,
+      "eval_spearmanr": 0.694155649948616,
+      "eval_steps_per_second": 4.24,
+      "step": 135
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 1.0297,
+      "step": 180
+    },
+    {
+      "epoch": 4.0,
+      "eval_combined_score": 0.6914969217648269,
+      "eval_loss": 1.6176120042800903,
+      "eval_pearson": 0.6869258091116714,
+      "eval_runtime": 2.8207,
+      "eval_samples_per_second": 531.79,
+      "eval_spearmanr": 0.6960680344179824,
+      "eval_steps_per_second": 4.254,
+      "step": 180
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.8064,
+      "step": 225
+    },
+    {
+      "epoch": 5.0,
+      "eval_combined_score": 0.7460100236019249,
+      "eval_loss": 1.1443842649459839,
+      "eval_pearson": 0.7475621408974276,
+      "eval_runtime": 2.8189,
+      "eval_samples_per_second": 532.114,
+      "eval_spearmanr": 0.7444579063064223,
+      "eval_steps_per_second": 4.257,
+      "step": 225
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 3.52e-05,
+      "loss": 0.604,
+      "step": 270
+    },
+    {
+      "epoch": 6.0,
+      "eval_combined_score": 0.7436314867221065,
+      "eval_loss": 1.2753756046295166,
+      "eval_pearson": 0.7422254000916392,
+      "eval_runtime": 2.8172,
+      "eval_samples_per_second": 532.449,
+      "eval_spearmanr": 0.7450375733525738,
+      "eval_steps_per_second": 4.26,
+      "step": 270
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 3.44e-05,
+      "loss": 0.4818,
+      "step": 315
+    },
+    {
+      "epoch": 7.0,
+      "eval_combined_score": 0.7680210248146195,
+      "eval_loss": 1.1406958103179932,
+      "eval_pearson": 0.7687318203548877,
+      "eval_runtime": 2.8308,
+      "eval_samples_per_second": 529.887,
+      "eval_spearmanr": 0.7673102292743514,
+      "eval_steps_per_second": 4.239,
+      "step": 315
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 0.3905,
+      "step": 360
+    },
+    {
+      "epoch": 8.0,
+      "eval_combined_score": 0.758227977038739,
+      "eval_loss": 1.1859891414642334,
+      "eval_pearson": 0.7560439754355848,
+      "eval_runtime": 2.8272,
+      "eval_samples_per_second": 530.555,
+      "eval_spearmanr": 0.7604119786418932,
+      "eval_steps_per_second": 4.244,
+      "step": 360
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 3.28e-05,
+      "loss": 0.3476,
+      "step": 405
+    },
+    {
+      "epoch": 9.0,
+      "eval_combined_score": 0.7493306108739174,
+      "eval_loss": 0.9799954295158386,
+      "eval_pearson": 0.7514687536537382,
+      "eval_runtime": 2.8292,
+      "eval_samples_per_second": 530.193,
+      "eval_spearmanr": 0.7471924680940966,
+      "eval_steps_per_second": 4.242,
+      "step": 405
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.2819,
+      "step": 450
+    },
+    {
+      "epoch": 10.0,
+      "eval_combined_score": 0.7514056262860007,
+      "eval_loss": 1.0156044960021973,
+      "eval_pearson": 0.7521360337612034,
+      "eval_runtime": 2.8179,
+      "eval_samples_per_second": 532.313,
+      "eval_spearmanr": 0.7506752188107981,
+      "eval_steps_per_second": 4.259,
+      "step": 450
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 3.1200000000000006e-05,
+      "loss": 0.2418,
+      "step": 495
+    },
+    {
+      "epoch": 11.0,
+      "eval_combined_score": 0.749784975779181,
+      "eval_loss": 1.0174038410186768,
+      "eval_pearson": 0.7516160448200117,
+      "eval_runtime": 2.8194,
+      "eval_samples_per_second": 532.024,
+      "eval_spearmanr": 0.7479539067383503,
+      "eval_steps_per_second": 4.256,
+      "step": 495
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 3.0400000000000004e-05,
+      "loss": 0.2068,
+      "step": 540
+    },
+    {
+      "epoch": 12.0,
+      "eval_combined_score": 0.7526651275891282,
+      "eval_loss": 1.236674427986145,
+      "eval_pearson": 0.7530062028197931,
+      "eval_runtime": 2.8215,
+      "eval_samples_per_second": 531.637,
+      "eval_spearmanr": 0.7523240523584632,
+      "eval_steps_per_second": 4.253,
+      "step": 540
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 2.96e-05,
+      "loss": 0.1863,
+      "step": 585
+    },
+    {
+      "epoch": 13.0,
+      "eval_combined_score": 0.7479613454649323,
+      "eval_loss": 1.0072920322418213,
+      "eval_pearson": 0.7491318790965454,
+      "eval_runtime": 2.8345,
+      "eval_samples_per_second": 529.2,
+      "eval_spearmanr": 0.7467908118333191,
+      "eval_steps_per_second": 4.234,
+      "step": 585
+    },
+    {
+      "epoch": 14.0,
+      "learning_rate": 2.8800000000000002e-05,
+      "loss": 0.1929,
+      "step": 630
+    },
+    {
+      "epoch": 14.0,
+      "eval_combined_score": 0.7511028300281887,
+      "eval_loss": 1.0470010042190552,
+      "eval_pearson": 0.7516884780316463,
+      "eval_runtime": 2.8162,
+      "eval_samples_per_second": 532.63,
+      "eval_spearmanr": 0.750517182024731,
+      "eval_steps_per_second": 4.261,
+      "step": 630
+    },
+    {
+      "epoch": 14.0,
+      "step": 630,
+      "total_flos": 1.1919621946867712e+16,
+      "train_loss": 0.7678726120600625,
+      "train_runtime": 566.4946,
+      "train_samples_per_second": 507.419,
+      "train_steps_per_second": 3.972
+    }
+  ],
+  "max_steps": 2250,
+  "num_train_epochs": 50,
+  "total_flos": 1.1919621946867712e+16,
+  "trial_name": null,
+  "trial_params": null
+}