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
@@ -20,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.4946000366099213
+      value: 0.5053999633900788
 ---
 
 <!-- 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_qnli
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2](https://huggingface.co/gokuls/bert_12_layer_model_v2) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2](https://huggingface.co/gokuls/bert_12_layer_model_v2) on the GLUE QNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6931
-- Accuracy: 0.4946
+- Loss: 0.6930
+- Accuracy: 0.5054
 
 ## Model description
 

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 13.0,
+    "eval_accuracy": 0.5053999633900788,
+    "eval_loss": 0.6929702162742615,
+    "eval_runtime": 6.4189,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 851.081,
+    "eval_steps_per_second": 3.427,
+    "train_loss": 0.6936567925601694,
+    "train_runtime": 4888.5149,
+    "train_samples": 104743,
+    "train_samples_per_second": 1071.317,
+    "train_steps_per_second": 4.194
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 13.0,
+    "eval_accuracy": 0.5053999633900788,
+    "eval_loss": 0.6929702162742615,
+    "eval_runtime": 6.4189,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 851.081,
+    "eval_steps_per_second": 3.427
+}

logs/events.out.tfevents.1677666330.kyoto.1072791.8

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f51eceecc57f82b2d01866a510409338649759e8958c78daff5df34fcc7db0f
+size 363

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 13.0,
+    "train_loss": 0.6936567925601694,
+    "train_runtime": 4888.5149,
+    "train_samples": 104743,
+    "train_samples_per_second": 1071.317,
+    "train_steps_per_second": 4.194
+}

trainer_state.json

@@ -0,0 +1,220 @@
+{
+  "best_metric": 0.6929702162742615,
+  "best_model_checkpoint": "hBERTv2_qnli/checkpoint-3280",
+  "epoch": 13.0,
+  "global_step": 5330,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.9e-05,
+      "loss": 0.6968,
+      "step": 410
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 0.6952407360076904,
+      "eval_runtime": 6.4563,
+      "eval_samples_per_second": 846.145,
+      "eval_steps_per_second": 3.408,
+      "step": 410
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.8e-05,
+      "loss": 0.6943,
+      "step": 820
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.4946000366099213,
+      "eval_loss": 0.6931500434875488,
+      "eval_runtime": 6.454,
+      "eval_samples_per_second": 846.448,
+      "eval_steps_per_second": 3.409,
+      "step": 820
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.7e-05,
+      "loss": 0.6937,
+      "step": 1230
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 0.6933012008666992,
+      "eval_runtime": 6.4678,
+      "eval_samples_per_second": 844.646,
+      "eval_steps_per_second": 3.401,
+      "step": 1230
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.6934,
+      "step": 1640
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 0.6930809617042542,
+      "eval_runtime": 6.4901,
+      "eval_samples_per_second": 841.738,
+      "eval_steps_per_second": 3.39,
+      "step": 1640
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 4.5e-05,
+      "loss": 0.6934,
+      "step": 2050
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 0.6931115388870239,
+      "eval_runtime": 6.4532,
+      "eval_samples_per_second": 846.551,
+      "eval_steps_per_second": 3.409,
+      "step": 2050
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.6933,
+      "step": 2460
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 0.6930229663848877,
+      "eval_runtime": 6.4635,
+      "eval_samples_per_second": 845.204,
+      "eval_steps_per_second": 3.404,
+      "step": 2460
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 4.3e-05,
+      "loss": 0.6933,
+      "step": 2870
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 0.6931431889533997,
+      "eval_runtime": 6.4426,
+      "eval_samples_per_second": 847.944,
+      "eval_steps_per_second": 3.415,
+      "step": 2870
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.2e-05,
+      "loss": 0.6932,
+      "step": 3280
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 0.6929702162742615,
+      "eval_runtime": 6.4464,
+      "eval_samples_per_second": 847.453,
+      "eval_steps_per_second": 3.413,
+      "step": 3280
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 4.1e-05,
+      "loss": 0.6932,
+      "step": 3690
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.4946000366099213,
+      "eval_loss": 0.6933699250221252,
+      "eval_runtime": 6.4536,
+      "eval_samples_per_second": 846.499,
+      "eval_steps_per_second": 3.409,
+      "step": 3690
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 4e-05,
+      "loss": 0.6932,
+      "step": 4100
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 0.6930176615715027,
+      "eval_runtime": 6.4577,
+      "eval_samples_per_second": 845.972,
+      "eval_steps_per_second": 3.407,
+      "step": 4100
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 3.9000000000000006e-05,
+      "loss": 0.6932,
+      "step": 4510
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.4946000366099213,
+      "eval_loss": 0.6931337118148804,
+      "eval_runtime": 6.4617,
+      "eval_samples_per_second": 845.439,
+      "eval_steps_per_second": 3.405,
+      "step": 4510
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 3.8e-05,
+      "loss": 0.6933,
+      "step": 4920
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.4946000366099213,
+      "eval_loss": 0.6934068202972412,
+      "eval_runtime": 6.4533,
+      "eval_samples_per_second": 846.539,
+      "eval_steps_per_second": 3.409,
+      "step": 4920
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 3.7e-05,
+      "loss": 0.6932,
+      "step": 5330
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.4946000366099213,
+      "eval_loss": 0.6931495070457458,
+      "eval_runtime": 6.4356,
+      "eval_samples_per_second": 848.877,
+      "eval_steps_per_second": 3.419,
+      "step": 5330
+    },
+    {
+      "epoch": 13.0,
+      "step": 5330,
+      "total_flos": 1.6959930780719514e+17,
+      "train_loss": 0.6936567925601694,
+      "train_runtime": 4888.5149,
+      "train_samples_per_second": 1071.317,
+      "train_steps_per_second": 4.194
+    }
+  ],
+  "max_steps": 20500,
+  "num_train_epochs": 50,
+  "total_flos": 1.6959930780719514e+17,
+  "trial_name": null,
+  "trial_params": null
+}