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: 0.3518535396581493
+      value: 0.20926356415783265
 ---
 
 <!-- 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_no_pretrain_stsb
 
-This model is a fine-tuned version of [](https://huggingface.co/) on the glue dataset.
+This model is a fine-tuned version of [](https://huggingface.co/) on the GLUE STSB dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.3053
-- Pearson: 0.3501
-- Spearmanr: 0.3519
-- Combined Score: 0.3510
+- Loss: 2.2174
+- Pearson: 0.1946
+- Spearmanr: 0.2093
+- Combined Score: 0.2019
 
 ## Model description
 

all_results.json

@@ -1,16 +1,16 @@
 {
-    "epoch": 6.0,
-    "eval_combined_score": NaN,
-    "eval_loss": 2.2499566078186035,
-    "eval_pearson": NaN,
-    "eval_runtime": 2.5119,
+    "epoch": 7.0,
+    "eval_combined_score": 0.20194085374544546,
+    "eval_loss": 2.2174315452575684,
+    "eval_pearson": 0.19461814333305827,
+    "eval_runtime": 6.5348,
     "eval_samples": 1500,
-    "eval_samples_per_second": 597.148,
-    "eval_spearmanr": NaN,
-    "eval_steps_per_second": 4.777,
-    "train_loss": 4.073463411684389,
-    "train_runtime": 238.0084,
+    "eval_samples_per_second": 229.54,
+    "eval_spearmanr": 0.20926356415783265,
+    "eval_steps_per_second": 1.836,
+    "train_loss": 1.4787501380557107,
+    "train_runtime": 694.6363,
     "train_samples": 5749,
-    "train_samples_per_second": 1207.731,
-    "train_steps_per_second": 9.453
+    "train_samples_per_second": 413.814,
+    "train_steps_per_second": 3.239
 }

eval_results.json

@@ -1,11 +1,11 @@
 {
-    "epoch": 6.0,
-    "eval_combined_score": NaN,
-    "eval_loss": 2.2499566078186035,
-    "eval_pearson": NaN,
-    "eval_runtime": 2.5119,
+    "epoch": 7.0,
+    "eval_combined_score": 0.20194085374544546,
+    "eval_loss": 2.2174315452575684,
+    "eval_pearson": 0.19461814333305827,
+    "eval_runtime": 6.5348,
     "eval_samples": 1500,
-    "eval_samples_per_second": 597.148,
-    "eval_spearmanr": NaN,
-    "eval_steps_per_second": 4.777
+    "eval_samples_per_second": 229.54,
+    "eval_spearmanr": 0.20926356415783265,
+    "eval_steps_per_second": 1.836
 }

logs/events.out.tfevents.1686799043.garda.2589545.13

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:20c5a7831fd4e61e75589ba10eceb9889cc1800c9f31deadaaddb99991e47375
+size 473

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 6.0,
-    "train_loss": 4.073463411684389,
-    "train_runtime": 238.0084,
+    "epoch": 7.0,
+    "train_loss": 1.4787501380557107,
+    "train_runtime": 694.6363,
     "train_samples": 5749,
-    "train_samples_per_second": 1207.731,
-    "train_steps_per_second": 9.453
+    "train_samples_per_second": 413.814,
+    "train_steps_per_second": 3.239
 }

trainer_state.json

@@ -1,127 +1,144 @@
 {
-  "best_metric": 2.2499566078186035,
-  "best_model_checkpoint": "hBERTv2_new_no_pretrain_stsb/checkpoint-45",
-  "epoch": 6.0,
-  "global_step": 270,
+  "best_metric": 2.2174315452575684,
+  "best_model_checkpoint": "hBERTv2_new_no_pretrain_stsb/checkpoint-90",
+  "epoch": 7.0,
+  "global_step": 315,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.0,
-      "learning_rate": 0.0004915555555555556,
-      "loss": 13.1642,
+      "learning_rate": 3.9200000000000004e-05,
+      "loss": 2.3893,
       "step": 45
     },
     {
       "epoch": 1.0,
-      "eval_combined_score": NaN,
-      "eval_loss": 2.2499566078186035,
-      "eval_pearson": NaN,
-      "eval_runtime": 2.5081,
-      "eval_samples_per_second": 598.074,
-      "eval_spearmanr": NaN,
-      "eval_steps_per_second": 4.785,
+      "eval_combined_score": 0.11711747086860938,
+      "eval_loss": 2.3697988986968994,
+      "eval_pearson": 0.12043925832454841,
+      "eval_runtime": 6.5953,
+      "eval_samples_per_second": 227.435,
+      "eval_spearmanr": 0.11379568341267035,
+      "eval_steps_per_second": 1.819,
       "step": 45
     },
     {
       "epoch": 2.0,
-      "learning_rate": 0.0004815555555555556,
-      "loss": 2.3438,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 1.9589,
       "step": 90
     },
     {
       "epoch": 2.0,
-      "eval_combined_score": -0.032927485831481466,
-      "eval_loss": 2.24997615814209,
-      "eval_pearson": -0.03356742396224663,
-      "eval_runtime": 2.5138,
-      "eval_samples_per_second": 596.712,
-      "eval_spearmanr": -0.032287547700716304,
-      "eval_steps_per_second": 4.774,
+      "eval_combined_score": 0.20194085374544546,
+      "eval_loss": 2.2174315452575684,
+      "eval_pearson": 0.19461814333305827,
+      "eval_runtime": 6.597,
+      "eval_samples_per_second": 227.377,
+      "eval_spearmanr": 0.20926356415783265,
+      "eval_steps_per_second": 1.819,
       "step": 90
     },
     {
       "epoch": 3.0,
-      "learning_rate": 0.0004715555555555556,
-      "loss": 2.3125,
+      "learning_rate": 3.76e-05,
+      "loss": 1.6743,
       "step": 135
     },
     {
       "epoch": 3.0,
-      "eval_combined_score": NaN,
-      "eval_loss": 2.250243663787842,
-      "eval_pearson": NaN,
-      "eval_runtime": 2.5223,
-      "eval_samples_per_second": 594.704,
-      "eval_spearmanr": NaN,
-      "eval_steps_per_second": 4.758,
+      "eval_combined_score": 0.2175435728704142,
+      "eval_loss": 2.3480913639068604,
+      "eval_pearson": 0.21439031391646174,
+      "eval_runtime": 6.5993,
+      "eval_samples_per_second": 227.299,
+      "eval_spearmanr": 0.22069683182436667,
+      "eval_steps_per_second": 1.818,
       "step": 135
     },
     {
       "epoch": 4.0,
-      "learning_rate": 0.00046155555555555555,
-      "loss": 2.2007,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 1.4068,
       "step": 180
     },
     {
       "epoch": 4.0,
-      "eval_combined_score": 0.044654595722518155,
-      "eval_loss": 2.3597922325134277,
-      "eval_pearson": 0.0455041521076301,
-      "eval_runtime": 2.5124,
-      "eval_samples_per_second": 597.043,
-      "eval_spearmanr": 0.04380503933740621,
-      "eval_steps_per_second": 4.776,
+      "eval_combined_score": 0.24955142048606804,
+      "eval_loss": 2.5921387672424316,
+      "eval_pearson": 0.247162444909398,
+      "eval_runtime": 6.6007,
+      "eval_samples_per_second": 227.248,
+      "eval_spearmanr": 0.2519403960627381,
+      "eval_steps_per_second": 1.818,
       "step": 180
     },
     {
       "epoch": 5.0,
-      "learning_rate": 0.0004515555555555556,
-      "loss": 2.2124,
+      "learning_rate": 3.6e-05,
+      "loss": 1.2205,
       "step": 225
     },
     {
       "epoch": 5.0,
-      "eval_combined_score": NaN,
-      "eval_loss": 2.3018898963928223,
-      "eval_pearson": NaN,
-      "eval_runtime": 2.5167,
-      "eval_samples_per_second": 596.027,
-      "eval_spearmanr": NaN,
-      "eval_steps_per_second": 4.768,
+      "eval_combined_score": 0.27093228753161847,
+      "eval_loss": 2.627875566482544,
+      "eval_pearson": 0.27178091382672603,
+      "eval_runtime": 6.595,
+      "eval_samples_per_second": 227.444,
+      "eval_spearmanr": 0.27008366123651084,
+      "eval_steps_per_second": 1.82,
       "step": 225
     },
     {
       "epoch": 6.0,
-      "learning_rate": 0.00044155555555555555,
-      "loss": 2.2073,
+      "learning_rate": 3.52e-05,
+      "loss": 0.9353,
       "step": 270
     },
     {
       "epoch": 6.0,
-      "eval_combined_score": NaN,
-      "eval_loss": 2.2925593852996826,
-      "eval_pearson": NaN,
-      "eval_runtime": 2.5136,
-      "eval_samples_per_second": 596.742,
-      "eval_spearmanr": NaN,
-      "eval_steps_per_second": 4.774,
+      "eval_combined_score": 0.316500979446309,
+      "eval_loss": 2.5439531803131104,
+      "eval_pearson": 0.3117440470832078,
+      "eval_runtime": 6.6054,
+      "eval_samples_per_second": 227.087,
+      "eval_spearmanr": 0.3212579118094102,
+      "eval_steps_per_second": 1.817,
       "step": 270
     },
     {
-      "epoch": 6.0,
-      "step": 270,
-      "total_flos": 5047782721191936.0,
-      "train_loss": 4.073463411684389,
-      "train_runtime": 238.0084,
-      "train_samples_per_second": 1207.731,
-      "train_steps_per_second": 9.453
+      "epoch": 7.0,
+      "learning_rate": 3.44e-05,
+      "loss": 0.7662,
+      "step": 315
+    },
+    {
+      "epoch": 7.0,
+      "eval_combined_score": 0.3509809608594302,
+      "eval_loss": 2.305265188217163,
+      "eval_pearson": 0.35010838206071104,
+      "eval_runtime": 6.5951,
+      "eval_samples_per_second": 227.443,
+      "eval_spearmanr": 0.3518535396581493,
+      "eval_steps_per_second": 1.82,
+      "step": 315
+    },
+    {
+      "epoch": 7.0,
+      "step": 315,
+      "total_flos": 5889079841390592.0,
+      "train_loss": 1.4787501380557107,
+      "train_runtime": 694.6363,
+      "train_samples_per_second": 413.814,
+      "train_steps_per_second": 3.239
     }
   ],
   "max_steps": 2250,
   "num_train_epochs": 50,
-  "total_flos": 5047782721191936.0,
+  "total_flos": 5889079841390592.0,
   "trial_name": null,
   "trial_params": null
 }