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 MNLI
       type: glue
       config: mnli
       split: validation_matched
@@ -20,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.31818644931227713
+      value: 0.3522172497965826
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -28,10 +30,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hBERTv2_new_no_pretrain_mnli
 
-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 MNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0986
-- Accuracy: 0.3182
+- Loss: 1.0983
+- Accuracy: 0.3522
 
 ## Model description
 

all_results.json

@@ -1,21 +1,21 @@
 {
-    "epoch": 11.0,
-    "epoch_mm": 11.0,
-    "eval_accuracy": 0.3273560876209883,
-    "eval_accuracy_mm": 0.3295362082994304,
-    "eval_loss": 1.0983835458755493,
-    "eval_loss_mm": 1.098366379737854,
-    "eval_runtime": 16.416,
-    "eval_runtime_mm": 16.4231,
+    "epoch": 7.0,
+    "epoch_mm": 7.0,
+    "eval_accuracy": 0.3544574630667346,
+    "eval_accuracy_mm": 0.3522172497965826,
+    "eval_loss": 1.0981956720352173,
+    "eval_loss_mm": 1.098296046257019,
+    "eval_runtime": 42.488,
+    "eval_runtime_mm": 42.7588,
     "eval_samples": 9815,
     "eval_samples_mm": 9832,
-    "eval_samples_per_second": 597.893,
-    "eval_samples_per_second_mm": 598.669,
-    "eval_steps_per_second": 4.691,
-    "eval_steps_per_second_mm": 4.689,
-    "train_loss": 1.099496340234514,
-    "train_runtime": 21144.6986,
+    "eval_samples_per_second": 231.006,
+    "eval_samples_per_second_mm": 229.941,
+    "eval_steps_per_second": 1.812,
+    "eval_steps_per_second_mm": 1.801,
+    "train_loss": 1.099199021855065,
+    "train_runtime": 36252.3204,
     "train_samples": 392702,
-    "train_samples_per_second": 928.606,
-    "train_steps_per_second": 7.255
+    "train_samples_per_second": 541.623,
+    "train_steps_per_second": 4.231
 }

eval_results.json

@@ -1,16 +1,16 @@
 {
-    "epoch": 11.0,
-    "epoch_mm": 11.0,
-    "eval_accuracy": 0.3273560876209883,
-    "eval_accuracy_mm": 0.3295362082994304,
-    "eval_loss": 1.0983835458755493,
-    "eval_loss_mm": 1.098366379737854,
-    "eval_runtime": 16.416,
-    "eval_runtime_mm": 16.4231,
+    "epoch": 7.0,
+    "epoch_mm": 7.0,
+    "eval_accuracy": 0.3544574630667346,
+    "eval_accuracy_mm": 0.3522172497965826,
+    "eval_loss": 1.0981956720352173,
+    "eval_loss_mm": 1.098296046257019,
+    "eval_runtime": 42.488,
+    "eval_runtime_mm": 42.7588,
     "eval_samples": 9815,
     "eval_samples_mm": 9832,
-    "eval_samples_per_second": 597.893,
-    "eval_samples_per_second_mm": 598.669,
-    "eval_steps_per_second": 4.691,
-    "eval_steps_per_second_mm": 4.689
+    "eval_samples_per_second": 231.006,
+    "eval_samples_per_second_mm": 229.941,
+    "eval_steps_per_second": 1.812,
+    "eval_steps_per_second_mm": 1.801
 }

logs/events.out.tfevents.1686836077.garda.2589545.17

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec3fab772e4ca227c4fbe1067facbdece401804c1e367f2510dbad208590e9c1
+size 698

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 11.0,
-    "train_loss": 1.099496340234514,
-    "train_runtime": 21144.6986,
+    "epoch": 7.0,
+    "train_loss": 1.099199021855065,
+    "train_runtime": 36252.3204,
     "train_samples": 392702,
-    "train_samples_per_second": 928.606,
-    "train_steps_per_second": 7.255
+    "train_samples_per_second": 541.623,
+    "train_steps_per_second": 4.231
 }

trainer_state.json

@@ -1,190 +1,130 @@
 {
-  "best_metric": 1.0983835458755493,
-  "best_model_checkpoint": "hBERTv2_new_no_pretrain_mnli/checkpoint-18408",
-  "epoch": 11.0,
-  "global_step": 33748,
+  "best_metric": 1.0981956720352173,
+  "best_model_checkpoint": "hBERTv2_new_no_pretrain_mnli/checkpoint-6136",
+  "epoch": 7.0,
+  "global_step": 21476,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.0,
-      "learning_rate": 0.000490009778357236,
-      "loss": 1.108,
+      "learning_rate": 3.9200000000000004e-05,
+      "loss": 1.1022,
       "step": 3068
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.3544574630667346,
-      "eval_loss": 1.0985444784164429,
-      "eval_runtime": 16.4465,
-      "eval_samples_per_second": 596.783,
-      "eval_steps_per_second": 4.682,
+      "eval_accuracy": 0.31818644931227713,
+      "eval_loss": 1.0985970497131348,
+      "eval_runtime": 42.8492,
+      "eval_samples_per_second": 229.059,
+      "eval_steps_per_second": 1.797,
       "step": 3068
     },
     {
       "epoch": 2.0,
-      "learning_rate": 0.00048000977835723597,
+      "learning_rate": 3.8400000000000005e-05,
       "loss": 1.0988,
       "step": 6136
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.31818644931227713,
-      "eval_loss": 1.0985891819000244,
-      "eval_runtime": 16.4454,
-      "eval_samples_per_second": 596.825,
-      "eval_steps_per_second": 4.682,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": 1.0981956720352173,
+      "eval_runtime": 42.7817,
+      "eval_samples_per_second": 229.421,
+      "eval_steps_per_second": 1.8,
       "step": 6136
     },
     {
       "epoch": 3.0,
-      "learning_rate": 0.000470009778357236,
-      "loss": 1.0986,
+      "learning_rate": 3.76e-05,
+      "loss": 1.0987,
       "step": 9204
     },
     {
       "epoch": 3.0,
       "eval_accuracy": 0.3273560876209883,
-      "eval_loss": 1.0984275341033936,
-      "eval_runtime": 16.4493,
-      "eval_samples_per_second": 596.681,
-      "eval_steps_per_second": 4.681,
+      "eval_loss": 1.0986144542694092,
+      "eval_runtime": 42.6833,
+      "eval_samples_per_second": 229.949,
+      "eval_steps_per_second": 1.804,
       "step": 9204
     },
     {
       "epoch": 4.0,
-      "learning_rate": 0.000460013037809648,
-      "loss": 1.0986,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 1.0988,
       "step": 12272
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.3544574630667346,
-      "eval_loss": 1.0988556146621704,
-      "eval_runtime": 16.4416,
-      "eval_samples_per_second": 596.962,
-      "eval_steps_per_second": 4.683,
+      "eval_accuracy": 0.31818644931227713,
+      "eval_loss": 1.0987833738327026,
+      "eval_runtime": 42.6772,
+      "eval_samples_per_second": 229.983,
+      "eval_steps_per_second": 1.804,
       "step": 12272
     },
     {
       "epoch": 5.0,
-      "learning_rate": 0.00045001629726206,
+      "learning_rate": 3.6e-05,
       "loss": 1.0986,
       "step": 15340
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.3544574630667346,
-      "eval_loss": 1.0984472036361694,
-      "eval_runtime": 16.4363,
-      "eval_samples_per_second": 597.153,
-      "eval_steps_per_second": 4.685,
+      "eval_accuracy": 0.3273560876209883,
+      "eval_loss": 1.0986013412475586,
+      "eval_runtime": 42.8795,
+      "eval_samples_per_second": 228.897,
+      "eval_steps_per_second": 1.796,
       "step": 15340
     },
     {
       "epoch": 6.0,
-      "learning_rate": 0.000440019556714472,
-      "loss": 1.0986,
+      "learning_rate": 3.52e-05,
+      "loss": 1.0987,
       "step": 18408
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.3273560876209883,
-      "eval_loss": 1.0983835458755493,
-      "eval_runtime": 16.345,
-      "eval_samples_per_second": 600.488,
-      "eval_steps_per_second": 4.711,
+      "eval_accuracy": 0.31818644931227713,
+      "eval_loss": 1.0986175537109375,
+      "eval_runtime": 42.7584,
+      "eval_samples_per_second": 229.546,
+      "eval_steps_per_second": 1.801,
       "step": 18408
     },
     {
       "epoch": 7.0,
-      "learning_rate": 0.00043002607561929595,
+      "learning_rate": 3.44e-05,
       "loss": 1.0986,
       "step": 21476
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.3273560876209883,
-      "eval_loss": 1.0984454154968262,
-      "eval_runtime": 16.3565,
-      "eval_samples_per_second": 600.068,
-      "eval_steps_per_second": 4.708,
-      "step": 21476
-    },
-    {
-      "epoch": 8.0,
-      "learning_rate": 0.000420029335071708,
-      "loss": 1.0986,
-      "step": 24544
-    },
-    {
-      "epoch": 8.0,
-      "eval_accuracy": 0.3544574630667346,
-      "eval_loss": 1.0985265970230103,
-      "eval_runtime": 16.438,
-      "eval_samples_per_second": 597.092,
-      "eval_steps_per_second": 4.684,
-      "step": 24544
-    },
-    {
-      "epoch": 9.0,
-      "learning_rate": 0.0004100358539765319,
-      "loss": 1.0986,
-      "step": 27612
-    },
-    {
-      "epoch": 9.0,
-      "eval_accuracy": 0.3544574630667346,
-      "eval_loss": 1.0984556674957275,
-      "eval_runtime": 16.4147,
-      "eval_samples_per_second": 597.938,
-      "eval_steps_per_second": 4.691,
-      "step": 27612
-    },
-    {
-      "epoch": 10.0,
-      "learning_rate": 0.00040003911342894395,
-      "loss": 1.0986,
-      "step": 30680
-    },
-    {
-      "epoch": 10.0,
-      "eval_accuracy": 0.31818644931227713,
-      "eval_loss": 1.098721981048584,
-      "eval_runtime": 16.3968,
-      "eval_samples_per_second": 598.593,
-      "eval_steps_per_second": 4.696,
-      "step": 30680
-    },
-    {
-      "epoch": 11.0,
-      "learning_rate": 0.0003900456323337679,
-      "loss": 1.0986,
-      "step": 33748
-    },
-    {
-      "epoch": 11.0,
       "eval_accuracy": 0.31818644931227713,
-      "eval_loss": 1.098907232284546,
-      "eval_runtime": 16.4429,
-      "eval_samples_per_second": 596.916,
-      "eval_steps_per_second": 4.683,
-      "step": 33748
+      "eval_loss": 1.098616600036621,
+      "eval_runtime": 42.7391,
+      "eval_samples_per_second": 229.649,
+      "eval_steps_per_second": 1.802,
+      "step": 21476
     },
     {
-      "epoch": 11.0,
-      "step": 33748,
-      "total_flos": 6.321496332708086e+17,
-      "train_loss": 1.099496340234514,
-      "train_runtime": 21144.6986,
-      "train_samples_per_second": 928.606,
-      "train_steps_per_second": 7.255
+      "epoch": 7.0,
+      "step": 21476,
+      "total_flos": 4.022770393541509e+17,
+      "train_loss": 1.099199021855065,
+      "train_runtime": 36252.3204,
+      "train_samples_per_second": 541.623,
+      "train_steps_per_second": 4.231
     }
   ],
   "max_steps": 153400,
   "num_train_epochs": 50,
-  "total_flos": 6.321496332708086e+17,
+  "total_flos": 4.022770393541509e+17,
   "trial_name": null,
   "trial_params": null
 }