End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 license: apache-2.0
 tags:
 - generated_from_trainer
@@ -13,7 +15,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE QNLI
       type: glue
       config: qnli
       split: validation
@@ -21,7 +23,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.7201171517481237
+      value: 0.6983342485813655
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -29,10 +31,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # sa_BERT_48_qnli
 
-This model is a fine-tuned version of [gokuls/bert_base_48](https://huggingface.co/gokuls/bert_base_48) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_base_48](https://huggingface.co/gokuls/bert_base_48) on the GLUE QNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.9555
-- Accuracy: 0.7201
+- Loss: 0.6317
+- Accuracy: 0.6983
 
 ## Model description
 

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 10.0,
+    "eval_accuracy": 0.6983342485813655,
+    "eval_loss": 0.6317353248596191,
+    "eval_runtime": 11.4148,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 478.59,
+    "eval_steps_per_second": 4.994,
+    "train_loss": 0.42643019288450806,
+    "train_runtime": 6922.6135,
+    "train_samples": 104743,
+    "train_samples_per_second": 756.528,
+    "train_steps_per_second": 7.887
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 10.0,
+    "eval_accuracy": 0.6983342485813655,
+    "eval_loss": 0.6317353248596191,
+    "eval_runtime": 11.4148,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 478.59,
+    "eval_steps_per_second": 4.994
+}

logs/events.out.tfevents.1686871331.serv-3318.1430967.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a671054660db1b74e0e9c7e783aae791863217b42b5de8ac5ffcf28a8373d026
+size 363

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 10.0,
+    "train_loss": 0.42643019288450806,
+    "train_runtime": 6922.6135,
+    "train_samples": 104743,
+    "train_samples_per_second": 756.528,
+    "train_steps_per_second": 7.887
+}

trainer_state.json

@@ -0,0 +1,175 @@
+{
+  "best_metric": 0.6317353248596191,
+  "best_model_checkpoint": "sa_BERT_48_qnli/checkpoint-5460",
+  "epoch": 10.0,
+  "global_step": 10920,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9200000000000004e-05,
+      "loss": 0.674,
+      "step": 1092
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6252974556104705,
+      "eval_loss": 0.6500131487846375,
+      "eval_runtime": 11.5317,
+      "eval_samples_per_second": 473.738,
+      "eval_steps_per_second": 4.943,
+      "step": 1092
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.6353,
+      "step": 2184
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6243822075782537,
+      "eval_loss": 0.6513252854347229,
+      "eval_runtime": 11.5355,
+      "eval_samples_per_second": 473.582,
+      "eval_steps_per_second": 4.941,
+      "step": 2184
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.76e-05,
+      "loss": 0.5987,
+      "step": 3276
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6357312831777412,
+      "eval_loss": 0.6552352905273438,
+      "eval_runtime": 11.5457,
+      "eval_samples_per_second": 473.165,
+      "eval_steps_per_second": 4.937,
+      "step": 3276
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.5429,
+      "step": 4368
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6760021965952773,
+      "eval_loss": 0.6413530111312866,
+      "eval_runtime": 11.5397,
+      "eval_samples_per_second": 473.409,
+      "eval_steps_per_second": 4.939,
+      "step": 4368
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.465,
+      "step": 5460
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6983342485813655,
+      "eval_loss": 0.6317353248596191,
+      "eval_runtime": 11.6176,
+      "eval_samples_per_second": 470.234,
+      "eval_steps_per_second": 4.906,
+      "step": 5460
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 3.52e-05,
+      "loss": 0.3904,
+      "step": 6552
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.7146256635548234,
+      "eval_loss": 0.6376231908798218,
+      "eval_runtime": 11.55,
+      "eval_samples_per_second": 472.985,
+      "eval_steps_per_second": 4.935,
+      "step": 6552
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 3.44e-05,
+      "loss": 0.3215,
+      "step": 7644
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.7137104155226066,
+      "eval_loss": 0.7152494192123413,
+      "eval_runtime": 11.5274,
+      "eval_samples_per_second": 473.914,
+      "eval_steps_per_second": 4.945,
+      "step": 7644
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 0.2584,
+      "step": 8736
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.7278052352187443,
+      "eval_loss": 0.769017219543457,
+      "eval_runtime": 11.5287,
+      "eval_samples_per_second": 473.862,
+      "eval_steps_per_second": 4.944,
+      "step": 8736
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 3.28e-05,
+      "loss": 0.2096,
+      "step": 9828
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.7127951674903898,
+      "eval_loss": 0.8507308959960938,
+      "eval_runtime": 11.5357,
+      "eval_samples_per_second": 473.575,
+      "eval_steps_per_second": 4.941,
+      "step": 9828
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.1685,
+      "step": 10920
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.7201171517481237,
+      "eval_loss": 0.955507755279541,
+      "eval_runtime": 11.5268,
+      "eval_samples_per_second": 473.94,
+      "eval_steps_per_second": 4.945,
+      "step": 10920
+    },
+    {
+      "epoch": 10.0,
+      "step": 10920,
+      "total_flos": 1.3779521027178496e+17,
+      "train_loss": 0.42643019288450806,
+      "train_runtime": 6922.6135,
+      "train_samples_per_second": 756.528,
+      "train_steps_per_second": 7.887
+    }
+  ],
+  "max_steps": 54600,
+  "num_train_epochs": 50,
+  "total_flos": 1.3779521027178496e+17,
+  "trial_name": null,
+  "trial_params": null
+}