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 QNLI
       type: glue
       config: qnli
       split: validation
@@ -28,7 +30,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hBERTv2_new_no_pretrain_qnli
 
-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 QNLI dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.6931
 - Accuracy: 0.5054

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.5053999633900788,
+    "eval_loss": 0.6930862069129944,
+    "eval_runtime": 9.1296,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 598.38,
+    "eval_steps_per_second": 4.71,
+    "train_loss": 0.697218982291309,
+    "train_runtime": 4274.4599,
+    "train_samples": 104743,
+    "train_samples_per_second": 1225.219,
+    "train_steps_per_second": 9.58
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.5053999633900788,
+    "eval_loss": 0.6930862069129944,
+    "eval_runtime": 9.1296,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 598.38,
+    "eval_steps_per_second": 4.71
+}

logs/events.out.tfevents.1685340481.serv-3317.2799588.8

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:270fd10cec4dbee11eadbb11602c4f984636eebada5399f937252e631e5d24e8
+size 363

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 8.0,
+    "train_loss": 0.697218982291309,
+    "train_runtime": 4274.4599,
+    "train_samples": 104743,
+    "train_samples_per_second": 1225.219,
+    "train_steps_per_second": 9.58
+}

trainer_state.json

@@ -0,0 +1,145 @@
+{
+  "best_metric": 0.6930862069129944,
+  "best_model_checkpoint": "hBERTv2_new_no_pretrain_qnli/checkpoint-2457",
+  "epoch": 8.0,
+  "global_step": 6552,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.00049002442002442,
+      "loss": 0.725,
+      "step": 819
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.4946000366099213,
+      "eval_loss": 0.6933751702308655,
+      "eval_runtime": 9.1662,
+      "eval_samples_per_second": 595.991,
+      "eval_steps_per_second": 4.691,
+      "step": 819
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 0.00048002442002442004,
+      "loss": 0.6932,
+      "step": 1638
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.4946000366099213,
+      "eval_loss": 0.6931600570678711,
+      "eval_runtime": 9.1659,
+      "eval_samples_per_second": 596.015,
+      "eval_steps_per_second": 4.691,
+      "step": 1638
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 0.00047002442002442,
+      "loss": 0.6932,
+      "step": 2457
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 0.6930862069129944,
+      "eval_runtime": 9.1571,
+      "eval_samples_per_second": 596.585,
+      "eval_steps_per_second": 4.696,
+      "step": 2457
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 0.00046003663003663006,
+      "loss": 0.6935,
+      "step": 3276
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.4946000366099213,
+      "eval_loss": 0.7160359025001526,
+      "eval_runtime": 9.1756,
+      "eval_samples_per_second": 595.383,
+      "eval_steps_per_second": 4.686,
+      "step": 3276
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 0.00045003663003663003,
+      "loss": 0.6935,
+      "step": 4095
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 0.6931073069572449,
+      "eval_runtime": 9.1507,
+      "eval_samples_per_second": 597.006,
+      "eval_steps_per_second": 4.699,
+      "step": 4095
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 0.00044003663003663,
+      "loss": 0.6932,
+      "step": 4914
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 0.6930915117263794,
+      "eval_runtime": 9.1533,
+      "eval_samples_per_second": 596.836,
+      "eval_steps_per_second": 4.698,
+      "step": 4914
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 0.0004300366300366301,
+      "loss": 0.6932,
+      "step": 5733
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 0.6931073069572449,
+      "eval_runtime": 9.1492,
+      "eval_samples_per_second": 597.098,
+      "eval_steps_per_second": 4.7,
+      "step": 5733
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 0.00042003663003663006,
+      "loss": 0.6932,
+      "step": 6552
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 0.6930862069129944,
+      "eval_runtime": 11.3762,
+      "eval_samples_per_second": 480.211,
+      "eval_steps_per_second": 3.78,
+      "step": 6552
+    },
+    {
+      "epoch": 8.0,
+      "step": 6552,
+      "total_flos": 1.2262403072013107e+17,
+      "train_loss": 0.697218982291309,
+      "train_runtime": 4274.4599,
+      "train_samples_per_second": 1225.219,
+      "train_steps_per_second": 9.58
+    }
+  ],
+  "max_steps": 40950,
+  "num_train_epochs": 50,
+  "total_flos": 1.2262403072013107e+17,
+  "trial_name": null,
+  "trial_params": null
+}