End of training

README.md

@@ -1,13 +1,28 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/tiny_bert_km_10_v1
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 model-index:
 - name: tiny_bert_km_10_v1_qnli
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE QNLI
+      type: glue
+      args: qnli
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.6326194398682042
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -15,10 +30,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # tiny_bert_km_10_v1_qnli
 
-This model is a fine-tuned version of [Hartunka/tiny_bert_km_10_v1](https://huggingface.co/Hartunka/tiny_bert_km_10_v1) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/tiny_bert_km_10_v1](https://huggingface.co/Hartunka/tiny_bert_km_10_v1) on the GLUE QNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0746
-- Accuracy: 0.6114
+- Loss: 0.6438
+- Accuracy: 0.6326
 
 ## Model description
 

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.6326194398682042,
+    "eval_loss": 0.6438024044036865,
+    "eval_runtime": 1.5914,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 3432.925,
+    "eval_steps_per_second": 13.825,
+    "total_flos": 2.197388993748173e+16,
+    "train_loss": 0.49335777236194145,
+    "train_runtime": 457.9004,
+    "train_samples": 104743,
+    "train_samples_per_second": 11437.312,
+    "train_steps_per_second": 44.77
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.6326194398682042,
+    "eval_loss": 0.6438024044036865,
+    "eval_runtime": 1.5914,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 3432.925,
+    "eval_steps_per_second": 13.825
+}

logs/events.out.tfevents.1744633505.s_005_m.2701049.23

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3bf8d2edaecd3bbdbd7d079e6c96f14b448cdd9c1d5593976e891597b683e599
+size 363

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 8.0,
+    "total_flos": 2.197388993748173e+16,
+    "train_loss": 0.49335777236194145,
+    "train_runtime": 457.9004,
+    "train_samples": 104743,
+    "train_samples_per_second": 11437.312,
+    "train_steps_per_second": 44.77
+}

trainer_state.json

@@ -0,0 +1,180 @@
+{
+  "best_global_step": 1230,
+  "best_metric": 0.6438024044036865,
+  "best_model_checkpoint": "tiny_bert_km_10_v1_qnli/checkpoint-1230",
+  "epoch": 8.0,
+  "eval_steps": 500,
+  "global_step": 3280,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.3384031057357788,
+      "learning_rate": 4.9e-05,
+      "loss": 0.6683,
+      "step": 410
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6174263225334066,
+      "eval_loss": 0.64871746301651,
+      "eval_runtime": 1.5971,
+      "eval_samples_per_second": 3420.51,
+      "eval_steps_per_second": 13.775,
+      "step": 410
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.429030179977417,
+      "learning_rate": 4.8e-05,
+      "loss": 0.6409,
+      "step": 820
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6231008603331503,
+      "eval_loss": 0.6487666964530945,
+      "eval_runtime": 1.5479,
+      "eval_samples_per_second": 3529.268,
+      "eval_steps_per_second": 14.213,
+      "step": 820
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 2.2790884971618652,
+      "learning_rate": 4.7e-05,
+      "loss": 0.6022,
+      "step": 1230
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6326194398682042,
+      "eval_loss": 0.6438024044036865,
+      "eval_runtime": 1.5551,
+      "eval_samples_per_second": 3512.935,
+      "eval_steps_per_second": 14.147,
+      "step": 1230
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 3.5781073570251465,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.5417,
+      "step": 1640
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6249313563975838,
+      "eval_loss": 0.6985598206520081,
+      "eval_runtime": 1.5497,
+      "eval_samples_per_second": 3525.244,
+      "eval_steps_per_second": 14.196,
+      "step": 1640
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 3.0873143672943115,
+      "learning_rate": 4.5e-05,
+      "loss": 0.4704,
+      "step": 2050
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6234669595460369,
+      "eval_loss": 0.7350184917449951,
+      "eval_runtime": 1.548,
+      "eval_samples_per_second": 3529.108,
+      "eval_steps_per_second": 14.212,
+      "step": 2050
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 5.652630805969238,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.4013,
+      "step": 2460
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6132161815852096,
+      "eval_loss": 0.8578618168830872,
+      "eval_runtime": 1.5465,
+      "eval_samples_per_second": 3532.584,
+      "eval_steps_per_second": 14.226,
+      "step": 2460
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 6.920729637145996,
+      "learning_rate": 4.3e-05,
+      "loss": 0.3381,
+      "step": 2870
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.599121361889072,
+      "eval_loss": 1.0881600379943848,
+      "eval_runtime": 1.5471,
+      "eval_samples_per_second": 3531.076,
+      "eval_steps_per_second": 14.22,
+      "step": 2870
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 4.917191505432129,
+      "learning_rate": 4.2e-05,
+      "loss": 0.2839,
+      "step": 3280
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6113856855207761,
+      "eval_loss": 1.0746463537216187,
+      "eval_runtime": 1.5511,
+      "eval_samples_per_second": 3521.915,
+      "eval_steps_per_second": 14.183,
+      "step": 3280
+    },
+    {
+      "epoch": 8.0,
+      "step": 3280,
+      "total_flos": 2.197388993748173e+16,
+      "train_loss": 0.49335777236194145,
+      "train_runtime": 457.9004,
+      "train_samples_per_second": 11437.312,
+      "train_steps_per_second": 44.77
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 20500,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 50,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 5,
+        "early_stopping_threshold": 0.0
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 5
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.197388993748173e+16,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}