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: -0.014776116124505849
+      value: 0.06376297357455894
 ---
 
 <!-- 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_new_pretrain_w_init__stsb
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_wt_init](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_wt_init) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_wt_init](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_wt_init) on the GLUE STSB dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.2834
-- Pearson: 0.0372
-- Spearmanr: -0.0148
-- Combined Score: 0.0112
+- Loss: 2.2578
+- Pearson: 0.0646
+- Spearmanr: 0.0638
+- Combined Score: 0.0642
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 8.0,
+    "eval_combined_score": 0.06419211108654013,
+    "eval_loss": 2.2578487396240234,
+    "eval_pearson": 0.06462124859852134,
+    "eval_runtime": 2.7153,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 552.426,
+    "eval_spearmanr": 0.06376297357455894,
+    "eval_steps_per_second": 4.419,
+    "train_loss": 9.75036686791314,
+    "train_runtime": 303.1246,
+    "train_samples": 5749,
+    "train_samples_per_second": 948.29,
+    "train_steps_per_second": 7.423
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 8.0,
+    "eval_combined_score": 0.06419211108654013,
+    "eval_loss": 2.2578487396240234,
+    "eval_pearson": 0.06462124859852134,
+    "eval_runtime": 2.7153,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 552.426,
+    "eval_spearmanr": 0.06376297357455894,
+    "eval_steps_per_second": 4.419
+}

logs/events.out.tfevents.1685552084.serv-9223.3032138.20

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:849b9169b3d2d383f548e0fed601a259c6725e8a2961b38f23e9699a9b5622c5
+size 473

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 8.0,
+    "train_loss": 9.75036686791314,
+    "train_runtime": 303.1246,
+    "train_samples": 5749,
+    "train_samples_per_second": 948.29,
+    "train_steps_per_second": 7.423
+}

trainer_state.json

@@ -0,0 +1,161 @@
+{
+  "best_metric": 2.2578487396240234,
+  "best_model_checkpoint": "hBERTv1_new_pretrain_w_init__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": 0.00049,
+      "loss": 61.8804,
+      "step": 45
+    },
+    {
+      "epoch": 1.0,
+      "eval_combined_score": -0.011402895045003632,
+      "eval_loss": 2.3511593341827393,
+      "eval_pearson": -0.043961204014640716,
+      "eval_runtime": 2.6935,
+      "eval_samples_per_second": 556.898,
+      "eval_spearmanr": 0.02115541392463345,
+      "eval_steps_per_second": 4.455,
+      "step": 45
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 0.00048,
+      "loss": 2.6713,
+      "step": 90
+    },
+    {
+      "epoch": 2.0,
+      "eval_combined_score": 0.030190925605361886,
+      "eval_loss": 3.0562541484832764,
+      "eval_pearson": 0.047891851367456564,
+      "eval_runtime": 2.705,
+      "eval_samples_per_second": 554.52,
+      "eval_spearmanr": 0.012489999843267211,
+      "eval_steps_per_second": 4.436,
+      "step": 90
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 0.00047,
+      "loss": 2.3803,
+      "step": 135
+    },
+    {
+      "epoch": 3.0,
+      "eval_combined_score": 0.06419211108654013,
+      "eval_loss": 2.2578487396240234,
+      "eval_pearson": 0.06462124859852134,
+      "eval_runtime": 2.702,
+      "eval_samples_per_second": 555.135,
+      "eval_spearmanr": 0.06376297357455894,
+      "eval_steps_per_second": 4.441,
+      "step": 135
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 0.00046,
+      "loss": 2.2516,
+      "step": 180
+    },
+    {
+      "epoch": 4.0,
+      "eval_combined_score": 0.06419211108654013,
+      "eval_loss": 2.3251523971557617,
+      "eval_pearson": 0.06462124859852134,
+      "eval_runtime": 2.7085,
+      "eval_samples_per_second": 553.814,
+      "eval_spearmanr": 0.06376297357455894,
+      "eval_steps_per_second": 4.431,
+      "step": 180
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 0.00045000000000000004,
+      "loss": 2.243,
+      "step": 225
+    },
+    {
+      "epoch": 5.0,
+      "eval_combined_score": 0.06419211108654015,
+      "eval_loss": 2.292773485183716,
+      "eval_pearson": 0.06462124859852135,
+      "eval_runtime": 2.71,
+      "eval_samples_per_second": 553.502,
+      "eval_spearmanr": 0.06376297357455894,
+      "eval_steps_per_second": 4.428,
+      "step": 225
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 0.00044,
+      "loss": 2.1764,
+      "step": 270
+    },
+    {
+      "epoch": 6.0,
+      "eval_combined_score": 0.03446148040040488,
+      "eval_loss": 2.282623529434204,
+      "eval_pearson": 0.06316650546297436,
+      "eval_runtime": 2.7071,
+      "eval_samples_per_second": 554.106,
+      "eval_spearmanr": 0.0057564553378353965,
+      "eval_steps_per_second": 4.433,
+      "step": 270
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 0.00043,
+      "loss": 2.1933,
+      "step": 315
+    },
+    {
+      "epoch": 7.0,
+      "eval_combined_score": 0.049598855935624034,
+      "eval_loss": 2.3737523555755615,
+      "eval_pearson": 0.04950619729973721,
+      "eval_runtime": 2.7069,
+      "eval_samples_per_second": 554.146,
+      "eval_spearmanr": 0.04969151457151086,
+      "eval_steps_per_second": 4.433,
+      "step": 315
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 0.00042,
+      "loss": 2.2067,
+      "step": 360
+    },
+    {
+      "epoch": 8.0,
+      "eval_combined_score": 0.011190672619407826,
+      "eval_loss": 2.2834489345550537,
+      "eval_pearson": 0.0371574613633215,
+      "eval_runtime": 2.714,
+      "eval_samples_per_second": 552.7,
+      "eval_spearmanr": -0.014776116124505849,
+      "eval_steps_per_second": 4.422,
+      "step": 360
+    },
+    {
+      "epoch": 8.0,
+      "step": 360,
+      "total_flos": 6811212541067264.0,
+      "train_loss": 9.75036686791314,
+      "train_runtime": 303.1246,
+      "train_samples_per_second": 948.29,
+      "train_steps_per_second": 7.423
+    }
+  ],
+  "max_steps": 2250,
+  "num_train_epochs": 50,
+  "total_flos": 6811212541067264.0,
+  "trial_name": null,
+  "trial_params": null
+}