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: 0.7823687091799434
+      value: 0.7706783096515127
 ---
 
 <!-- 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_stsb
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2](https://huggingface.co/gokuls/bert_12_layer_model_v2) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2](https://huggingface.co/gokuls/bert_12_layer_model_v2) on the GLUE STSB dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.1525
-- Pearson: 0.7836
-- Spearmanr: 0.7824
-- Combined Score: 0.7830
+- Loss: 0.9534
+- Pearson: 0.7722
+- Spearmanr: 0.7707
+- Combined Score: 0.7714
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 10.0,
+    "eval_combined_score": 0.7714202291712986,
+    "eval_loss": 0.9534441828727722,
+    "eval_pearson": 0.7721621486910845,
+    "eval_runtime": 1.7529,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 855.716,
+    "eval_spearmanr": 0.7706783096515127,
+    "eval_steps_per_second": 3.423,
+    "train_loss": 1.0605290993400243,
+    "train_runtime": 262.8334,
+    "train_samples": 5749,
+    "train_samples_per_second": 1093.659,
+    "train_steps_per_second": 4.375
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 10.0,
+    "eval_combined_score": 0.7714202291712986,
+    "eval_loss": 0.9534441828727722,
+    "eval_pearson": 0.7721621486910845,
+    "eval_runtime": 1.7529,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 855.716,
+    "eval_spearmanr": 0.7706783096515127,
+    "eval_steps_per_second": 3.423
+}

logs/events.out.tfevents.1677758213.serv-3317.65116.20

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:50cacdf11df3d709231a871d9d3d3098c5f8281fc0b02a46e86768fc0aa7f6b0
+size 473

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 10.0,
+    "train_loss": 1.0605290993400243,
+    "train_runtime": 262.8334,
+    "train_samples": 5749,
+    "train_samples_per_second": 1093.659,
+    "train_steps_per_second": 4.375
+}

trainer_state.json

@@ -0,0 +1,195 @@
+{
+  "best_metric": 0.9534441828727722,
+  "best_model_checkpoint": "hBERTv2_stsb/checkpoint-115",
+  "epoch": 10.0,
+  "global_step": 230,
+  "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": 4.4386,
+      "step": 23
+    },
+    {
+      "epoch": 1.0,
+      "eval_combined_score": 0.11921757894035995,
+      "eval_loss": 2.5331363677978516,
+      "eval_pearson": 0.13130549535273073,
+      "eval_runtime": 1.7547,
+      "eval_samples_per_second": 854.84,
+      "eval_spearmanr": 0.10712966252798917,
+      "eval_steps_per_second": 3.419,
+      "step": 23
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.817391304347826e-05,
+      "loss": 1.8741,
+      "step": 46
+    },
+    {
+      "epoch": 2.0,
+      "eval_combined_score": 0.48443203302065396,
+      "eval_loss": 2.0517449378967285,
+      "eval_pearson": 0.4922986407408125,
+      "eval_runtime": 1.7533,
+      "eval_samples_per_second": 855.528,
+      "eval_spearmanr": 0.4765654253004954,
+      "eval_steps_per_second": 3.422,
+      "step": 46
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7173913043478264e-05,
+      "loss": 1.347,
+      "step": 69
+    },
+    {
+      "epoch": 3.0,
+      "eval_combined_score": 0.7021926806891676,
+      "eval_loss": 1.3556339740753174,
+      "eval_pearson": 0.6964421878346847,
+      "eval_runtime": 1.7555,
+      "eval_samples_per_second": 854.451,
+      "eval_spearmanr": 0.7079431735436507,
+      "eval_steps_per_second": 3.418,
+      "step": 69
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.617391304347826e-05,
+      "loss": 0.8443,
+      "step": 92
+    },
+    {
+      "epoch": 4.0,
+      "eval_combined_score": 0.7353483613374537,
+      "eval_loss": 1.2583037614822388,
+      "eval_pearson": 0.7339982958163664,
+      "eval_runtime": 1.7466,
+      "eval_samples_per_second": 858.817,
+      "eval_spearmanr": 0.736698426858541,
+      "eval_steps_per_second": 3.435,
+      "step": 92
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.517391304347826e-05,
+      "loss": 0.5822,
+      "step": 115
+    },
+    {
+      "epoch": 5.0,
+      "eval_combined_score": 0.7714202291712986,
+      "eval_loss": 0.9534441828727722,
+      "eval_pearson": 0.7721621486910845,
+      "eval_runtime": 1.7425,
+      "eval_samples_per_second": 860.836,
+      "eval_spearmanr": 0.7706783096515127,
+      "eval_steps_per_second": 3.443,
+      "step": 115
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4173913043478265e-05,
+      "loss": 0.4356,
+      "step": 138
+    },
+    {
+      "epoch": 6.0,
+      "eval_combined_score": 0.7784518955721378,
+      "eval_loss": 1.19211745262146,
+      "eval_pearson": 0.7797837440863491,
+      "eval_runtime": 1.7549,
+      "eval_samples_per_second": 854.75,
+      "eval_spearmanr": 0.7771200470579264,
+      "eval_steps_per_second": 3.419,
+      "step": 138
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.317391304347827e-05,
+      "loss": 0.3531,
+      "step": 161
+    },
+    {
+      "epoch": 7.0,
+      "eval_combined_score": 0.7700287306666532,
+      "eval_loss": 1.3848626613616943,
+      "eval_pearson": 0.7700627816034332,
+      "eval_runtime": 1.749,
+      "eval_samples_per_second": 857.639,
+      "eval_spearmanr": 0.7699946797298732,
+      "eval_steps_per_second": 3.431,
+      "step": 161
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.2173913043478264e-05,
+      "loss": 0.2712,
+      "step": 184
+    },
+    {
+      "epoch": 8.0,
+      "eval_combined_score": 0.7877855529630069,
+      "eval_loss": 1.0015486478805542,
+      "eval_pearson": 0.7885967294161835,
+      "eval_runtime": 1.7474,
+      "eval_samples_per_second": 858.407,
+      "eval_spearmanr": 0.7869743765098304,
+      "eval_steps_per_second": 3.434,
+      "step": 184
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 4.117391304347826e-05,
+      "loss": 0.259,
+      "step": 207
+    },
+    {
+      "epoch": 9.0,
+      "eval_combined_score": 0.7885662946075902,
+      "eval_loss": 1.0522509813308716,
+      "eval_pearson": 0.7897669906981138,
+      "eval_runtime": 1.7561,
+      "eval_samples_per_second": 854.151,
+      "eval_spearmanr": 0.7873655985170667,
+      "eval_steps_per_second": 3.417,
+      "step": 207
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 4.017391304347826e-05,
+      "loss": 0.2003,
+      "step": 230
+    },
+    {
+      "epoch": 10.0,
+      "eval_combined_score": 0.7829928810160273,
+      "eval_loss": 1.1524564027786255,
+      "eval_pearson": 0.7836170528521111,
+      "eval_runtime": 1.7407,
+      "eval_samples_per_second": 861.726,
+      "eval_spearmanr": 0.7823687091799434,
+      "eval_steps_per_second": 3.447,
+      "step": 230
+    },
+    {
+      "epoch": 10.0,
+      "step": 230,
+      "total_flos": 7160509077913600.0,
+      "train_loss": 1.0605290993400243,
+      "train_runtime": 262.8334,
+      "train_samples_per_second": 1093.659,
+      "train_steps_per_second": 4.375
+    }
+  ],
+  "max_steps": 1150,
+  "num_train_epochs": 50,
+  "total_flos": 7160509077913600.0,
+  "trial_name": null,
+  "trial_params": null
+}