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: 0.6941784502109372
+      value: 0.7155715863961268
 ---
 
 <!-- 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_stsb
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1](https://huggingface.co/gokuls/bert_12_layer_model_v1) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1](https://huggingface.co/gokuls/bert_12_layer_model_v1) on the GLUE STSB dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.2614
-- Pearson: 0.6982
-- Spearmanr: 0.6942
-- Combined Score: 0.6962
+- Loss: 1.1154
+- Pearson: 0.7159
+- Spearmanr: 0.7156
+- Combined Score: 0.7157
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 13.0,
+    "eval_combined_score": 0.7157490703420486,
+    "eval_loss": 1.1154277324676514,
+    "eval_pearson": 0.7159265542879704,
+    "eval_runtime": 1.917,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 782.476,
+    "eval_spearmanr": 0.7155715863961268,
+    "eval_steps_per_second": 3.13,
+    "train_loss": 1.0456982304818654,
+    "train_runtime": 396.42,
+    "train_samples": 5749,
+    "train_samples_per_second": 725.115,
+    "train_steps_per_second": 2.901
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 13.0,
+    "eval_combined_score": 0.7157490703420486,
+    "eval_loss": 1.1154277324676514,
+    "eval_pearson": 0.7159265542879704,
+    "eval_runtime": 1.917,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 782.476,
+    "eval_spearmanr": 0.7155715863961268,
+    "eval_steps_per_second": 3.13
+}

logs/events.out.tfevents.1677756168.serv-9223.249682.20

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc888a54b43ca33217531aa62a6610a9aabf41e433d1abf72452883ab4f6bb4d
+size 473

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 13.0,
+    "train_loss": 1.0456982304818654,
+    "train_runtime": 396.42,
+    "train_samples": 5749,
+    "train_samples_per_second": 725.115,
+    "train_steps_per_second": 2.901
+}

trainer_state.json

@@ -0,0 +1,246 @@
+{
+  "best_metric": 1.1154277324676514,
+  "best_model_checkpoint": "hBERTv1_stsb/checkpoint-184",
+  "epoch": 13.0,
+  "global_step": 299,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.9173913043478265e-05,
+      "loss": 4.0796,
+      "step": 23
+    },
+    {
+      "epoch": 1.0,
+      "eval_combined_score": 0.06538801707026302,
+      "eval_loss": 2.301738977432251,
+      "eval_pearson": 0.07610945358839591,
+      "eval_runtime": 1.9306,
+      "eval_samples_per_second": 776.964,
+      "eval_spearmanr": 0.05466658055213014,
+      "eval_steps_per_second": 3.108,
+      "step": 23
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.817391304347826e-05,
+      "loss": 2.0746,
+      "step": 46
+    },
+    {
+      "epoch": 2.0,
+      "eval_combined_score": 0.08111168068438418,
+      "eval_loss": 2.618116855621338,
+      "eval_pearson": 0.08503304117233579,
+      "eval_runtime": 1.9212,
+      "eval_samples_per_second": 780.765,
+      "eval_spearmanr": 0.07719032019643259,
+      "eval_steps_per_second": 3.123,
+      "step": 46
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7173913043478264e-05,
+      "loss": 1.9142,
+      "step": 69
+    },
+    {
+      "epoch": 3.0,
+      "eval_combined_score": 0.18653024063604068,
+      "eval_loss": 2.29634952545166,
+      "eval_pearson": 0.1878248631380315,
+      "eval_runtime": 1.9303,
+      "eval_samples_per_second": 777.095,
+      "eval_spearmanr": 0.18523561813404985,
+      "eval_steps_per_second": 3.108,
+      "step": 69
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.617391304347826e-05,
+      "loss": 1.6883,
+      "step": 92
+    },
+    {
+      "epoch": 4.0,
+      "eval_combined_score": 0.47587247941695476,
+      "eval_loss": 2.18662691116333,
+      "eval_pearson": 0.4740108701640409,
+      "eval_runtime": 1.9255,
+      "eval_samples_per_second": 779.032,
+      "eval_spearmanr": 0.4777340886698686,
+      "eval_steps_per_second": 3.116,
+      "step": 92
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.517391304347826e-05,
+      "loss": 1.1166,
+      "step": 115
+    },
+    {
+      "epoch": 5.0,
+      "eval_combined_score": 0.6384359218484894,
+      "eval_loss": 1.9366592168807983,
+      "eval_pearson": 0.6318718203898596,
+      "eval_runtime": 1.9201,
+      "eval_samples_per_second": 781.199,
+      "eval_spearmanr": 0.6450000233071193,
+      "eval_steps_per_second": 3.125,
+      "step": 115
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4173913043478265e-05,
+      "loss": 0.7598,
+      "step": 138
+    },
+    {
+      "epoch": 6.0,
+      "eval_combined_score": 0.684468485485146,
+      "eval_loss": 1.4188166856765747,
+      "eval_pearson": 0.6801003203440246,
+      "eval_runtime": 1.9182,
+      "eval_samples_per_second": 781.975,
+      "eval_spearmanr": 0.6888366506262675,
+      "eval_steps_per_second": 3.128,
+      "step": 138
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.317391304347827e-05,
+      "loss": 0.5453,
+      "step": 161
+    },
+    {
+      "epoch": 7.0,
+      "eval_combined_score": 0.6994286994310785,
+      "eval_loss": 1.2720448970794678,
+      "eval_pearson": 0.6987976257082615,
+      "eval_runtime": 1.9166,
+      "eval_samples_per_second": 782.626,
+      "eval_spearmanr": 0.7000597731538957,
+      "eval_steps_per_second": 3.131,
+      "step": 161
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.2173913043478264e-05,
+      "loss": 0.3705,
+      "step": 184
+    },
+    {
+      "epoch": 8.0,
+      "eval_combined_score": 0.7157490703420486,
+      "eval_loss": 1.1154277324676514,
+      "eval_pearson": 0.7159265542879704,
+      "eval_runtime": 1.9188,
+      "eval_samples_per_second": 781.732,
+      "eval_spearmanr": 0.7155715863961268,
+      "eval_steps_per_second": 3.127,
+      "step": 184
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 4.117391304347826e-05,
+      "loss": 0.2976,
+      "step": 207
+    },
+    {
+      "epoch": 9.0,
+      "eval_combined_score": 0.6780274177666261,
+      "eval_loss": 1.6889050006866455,
+      "eval_pearson": 0.6753593078005974,
+      "eval_runtime": 1.9186,
+      "eval_samples_per_second": 781.809,
+      "eval_spearmanr": 0.680695527732655,
+      "eval_steps_per_second": 3.127,
+      "step": 207
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 4.017391304347826e-05,
+      "loss": 0.2272,
+      "step": 230
+    },
+    {
+      "epoch": 10.0,
+      "eval_combined_score": 0.691436341571948,
+      "eval_loss": 1.3627489805221558,
+      "eval_pearson": 0.692937129271751,
+      "eval_runtime": 1.9208,
+      "eval_samples_per_second": 780.912,
+      "eval_spearmanr": 0.6899355538721449,
+      "eval_steps_per_second": 3.124,
+      "step": 230
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 3.917391304347826e-05,
+      "loss": 0.1966,
+      "step": 253
+    },
+    {
+      "epoch": 11.0,
+      "eval_combined_score": 0.7181059118945743,
+      "eval_loss": 1.127779245376587,
+      "eval_pearson": 0.7195047805569822,
+      "eval_runtime": 1.9246,
+      "eval_samples_per_second": 779.363,
+      "eval_spearmanr": 0.7167070432321663,
+      "eval_steps_per_second": 3.117,
+      "step": 253
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 3.817391304347827e-05,
+      "loss": 0.1708,
+      "step": 276
+    },
+    {
+      "epoch": 12.0,
+      "eval_combined_score": 0.7167948198834158,
+      "eval_loss": 1.3475854396820068,
+      "eval_pearson": 0.717074708327728,
+      "eval_runtime": 1.9199,
+      "eval_samples_per_second": 781.302,
+      "eval_spearmanr": 0.7165149314391035,
+      "eval_steps_per_second": 3.125,
+      "step": 276
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 3.7173913043478264e-05,
+      "loss": 0.1529,
+      "step": 299
+    },
+    {
+      "epoch": 13.0,
+      "eval_combined_score": 0.6961706036044664,
+      "eval_loss": 1.26143217086792,
+      "eval_pearson": 0.6981627569979957,
+      "eval_runtime": 1.9277,
+      "eval_samples_per_second": 778.118,
+      "eval_spearmanr": 0.6941784502109372,
+      "eval_steps_per_second": 3.112,
+      "step": 299
+    },
+    {
+      "epoch": 13.0,
+      "step": 299,
+      "total_flos": 9440019617939456.0,
+      "train_loss": 1.0456982304818654,
+      "train_runtime": 396.42,
+      "train_samples_per_second": 725.115,
+      "train_steps_per_second": 2.901
+    }
+  ],
+  "max_steps": 1150,
+  "num_train_epochs": 50,
+  "total_flos": 9440019617939456.0,
+  "trial_name": null,
+  "trial_params": null
+}