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: 0.283272028610209
+      value: 0.25026626408652064
 ---
 
 <!-- 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. -->
 
 # hBERTv2_new_pretrain_w_init_48_stsb
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new_wt_init_48](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new_wt_init_48) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new_wt_init_48](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new_wt_init_48) on the GLUE STSB dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.6252
-- Pearson: 0.2837
-- Spearmanr: 0.2833
-- Combined Score: 0.2835
+- Loss: 2.2569
+- Pearson: 0.2426
+- Spearmanr: 0.2503
+- Combined Score: 0.2465
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 8.0,
+    "eval_combined_score": 0.24645036793946712,
+    "eval_loss": 2.2569007873535156,
+    "eval_pearson": 0.2426344717924136,
+    "eval_runtime": 2.6063,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 575.522,
+    "eval_spearmanr": 0.25026626408652064,
+    "eval_steps_per_second": 4.604,
+    "train_loss": 1.3476022349463568,
+    "train_runtime": 308.5467,
+    "train_samples": 5749,
+    "train_samples_per_second": 931.626,
+    "train_steps_per_second": 7.292
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 8.0,
+    "eval_combined_score": 0.24645036793946712,
+    "eval_loss": 2.2569007873535156,
+    "eval_pearson": 0.2426344717924136,
+    "eval_runtime": 2.6063,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 575.522,
+    "eval_spearmanr": 0.25026626408652064,
+    "eval_steps_per_second": 4.604
+}

logs/events.out.tfevents.1686059902.serv-9207.2390652.20

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd70fb4433743d277d895f7a0923261387577f1396e2135c8ff175f0bc189741
+size 473

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 8.0,
+    "train_loss": 1.3476022349463568,
+    "train_runtime": 308.5467,
+    "train_samples": 5749,
+    "train_samples_per_second": 931.626,
+    "train_steps_per_second": 7.292
+}

trainer_state.json

@@ -0,0 +1,161 @@
+{
+  "best_metric": 2.2569007873535156,
+  "best_model_checkpoint": "hBERTv2_new_pretrain_w_init_48_stsb/checkpoint-135",
+  "epoch": 8.0,
+  "global_step": 360,
+  "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": 2.3064,
+      "step": 45
+    },
+    {
+      "epoch": 1.0,
+      "eval_combined_score": 0.1462439839108577,
+      "eval_loss": 2.2783970832824707,
+      "eval_pearson": 0.14921550721008922,
+      "eval_runtime": 2.6216,
+      "eval_samples_per_second": 572.162,
+      "eval_spearmanr": 0.14327246061162616,
+      "eval_steps_per_second": 4.577,
+      "step": 45
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 1.9454,
+      "step": 90
+    },
+    {
+      "epoch": 2.0,
+      "eval_combined_score": 0.26030447251939637,
+      "eval_loss": 2.2772936820983887,
+      "eval_pearson": 0.26598543333404057,
+      "eval_runtime": 2.6525,
+      "eval_samples_per_second": 565.495,
+      "eval_spearmanr": 0.2546235117047522,
+      "eval_steps_per_second": 4.524,
+      "step": 90
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.76e-05,
+      "loss": 1.7733,
+      "step": 135
+    },
+    {
+      "epoch": 3.0,
+      "eval_combined_score": 0.24645036793946712,
+      "eval_loss": 2.2569007873535156,
+      "eval_pearson": 0.2426344717924136,
+      "eval_runtime": 2.6404,
+      "eval_samples_per_second": 568.092,
+      "eval_spearmanr": 0.25026626408652064,
+      "eval_steps_per_second": 4.545,
+      "step": 135
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 1.4752,
+      "step": 180
+    },
+    {
+      "epoch": 4.0,
+      "eval_combined_score": 0.29290243765316837,
+      "eval_loss": 2.3394598960876465,
+      "eval_pearson": 0.2832887660979118,
+      "eval_runtime": 2.6479,
+      "eval_samples_per_second": 566.491,
+      "eval_spearmanr": 0.30251610920842487,
+      "eval_steps_per_second": 4.532,
+      "step": 180
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.6e-05,
+      "loss": 1.083,
+      "step": 225
+    },
+    {
+      "epoch": 5.0,
+      "eval_combined_score": 0.30418757127487117,
+      "eval_loss": 2.4139511585235596,
+      "eval_pearson": 0.30172563341382985,
+      "eval_runtime": 2.6618,
+      "eval_samples_per_second": 563.529,
+      "eval_spearmanr": 0.30664950913591255,
+      "eval_steps_per_second": 4.508,
+      "step": 225
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 3.52e-05,
+      "loss": 0.8694,
+      "step": 270
+    },
+    {
+      "epoch": 6.0,
+      "eval_combined_score": 0.29075556065228314,
+      "eval_loss": 2.885422468185425,
+      "eval_pearson": 0.27900766009179806,
+      "eval_runtime": 2.6377,
+      "eval_samples_per_second": 568.667,
+      "eval_spearmanr": 0.30250346121276817,
+      "eval_steps_per_second": 4.549,
+      "step": 270
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 3.44e-05,
+      "loss": 0.7272,
+      "step": 315
+    },
+    {
+      "epoch": 7.0,
+      "eval_combined_score": 0.30536562210653795,
+      "eval_loss": 2.890129566192627,
+      "eval_pearson": 0.30412357536912293,
+      "eval_runtime": 2.6421,
+      "eval_samples_per_second": 567.735,
+      "eval_spearmanr": 0.3066076688439529,
+      "eval_steps_per_second": 4.542,
+      "step": 315
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 0.6008,
+      "step": 360
+    },
+    {
+      "epoch": 8.0,
+      "eval_combined_score": 0.283502287603587,
+      "eval_loss": 2.6252315044403076,
+      "eval_pearson": 0.283732546596965,
+      "eval_runtime": 2.6457,
+      "eval_samples_per_second": 566.952,
+      "eval_spearmanr": 0.283272028610209,
+      "eval_steps_per_second": 4.536,
+      "step": 360
+    },
+    {
+      "epoch": 8.0,
+      "step": 360,
+      "total_flos": 6730376961589248.0,
+      "train_loss": 1.3476022349463568,
+      "train_runtime": 308.5467,
+      "train_samples_per_second": 931.626,
+      "train_steps_per_second": 7.292
+    }
+  ],
+  "max_steps": 2250,
+  "num_train_epochs": 50,
+  "total_flos": 6730376961589248.0,
+  "trial_name": null,
+  "trial_params": null
+}