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
@@ -28,9 +30,9 @@ 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: 2.2926
+- Loss: 2.2500
 - Pearson: nan
 - Spearmanr: nan
 - Combined Score: nan

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 6.0,
+    "eval_combined_score": NaN,
+    "eval_loss": 2.2499566078186035,
+    "eval_pearson": NaN,
+    "eval_runtime": 2.5119,
+    "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,
+    "train_samples": 5749,
+    "train_samples_per_second": 1207.731,
+    "train_steps_per_second": 9.453
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 6.0,
+    "eval_combined_score": NaN,
+    "eval_loss": 2.2499566078186035,
+    "eval_pearson": NaN,
+    "eval_runtime": 2.5119,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 597.148,
+    "eval_spearmanr": NaN,
+    "eval_steps_per_second": 4.777
+}

logs/events.out.tfevents.1685362584.serv-3317.2799588.20

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6d0cf436fbb063a1b43bc5d1f8a11505f4f95bd13cbd3d4c1bdfb1fb94e458b0
+size 473

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 6.0,
+    "train_loss": 4.073463411684389,
+    "train_runtime": 238.0084,
+    "train_samples": 5749,
+    "train_samples_per_second": 1207.731,
+    "train_steps_per_second": 9.453
+}

trainer_state.json

@@ -0,0 +1,127 @@
+{
+  "best_metric": 2.2499566078186035,
+  "best_model_checkpoint": "hBERTv2_new_no_pretrain_stsb/checkpoint-45",
+  "epoch": 6.0,
+  "global_step": 270,
+  "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,
+      "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,
+      "step": 45
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 0.0004815555555555556,
+      "loss": 2.3438,
+      "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,
+      "step": 90
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 0.0004715555555555556,
+      "loss": 2.3125,
+      "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,
+      "step": 135
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 0.00046155555555555555,
+      "loss": 2.2007,
+      "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,
+      "step": 180
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 0.0004515555555555556,
+      "loss": 2.2124,
+      "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,
+      "step": 225
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 0.00044155555555555555,
+      "loss": 2.2073,
+      "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,
+      "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
+    }
+  ],
+  "max_steps": 2250,
+  "num_train_epochs": 50,
+  "total_flos": 5047782721191936.0,
+  "trial_name": null,
+  "trial_params": null
+}