End of training

README.md

@@ -1,13 +1,28 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/tiny_bert_km_20_v2
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 model-index:
 - name: tiny_bert_km_20_v2_qnli
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE QNLI
+      type: glue
+      args: qnli
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.6278601501006773
 ---
 
 <!-- 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_20_v2_qnli
 
-This model is a fine-tuned version of [Hartunka/tiny_bert_km_20_v2](https://huggingface.co/Hartunka/tiny_bert_km_20_v2) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/tiny_bert_km_20_v2](https://huggingface.co/Hartunka/tiny_bert_km_20_v2) on the GLUE QNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.1298
-- Accuracy: 0.6077
+- Loss: 0.6494
+- Accuracy: 0.6279
 
 ## Model description
 

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.6278601501006773,
+    "eval_loss": 0.6493877172470093,
+    "eval_runtime": 1.5778,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 3462.394,
+    "eval_steps_per_second": 13.943,
+    "total_flos": 2.197388993748173e+16,
+    "train_loss": 0.4819710405861459,
+    "train_runtime": 453.5458,
+    "train_samples": 104743,
+    "train_samples_per_second": 11547.124,
+    "train_steps_per_second": 45.199
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.6278601501006773,
+    "eval_loss": 0.6493877172470093,
+    "eval_runtime": 1.5778,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 3462.394,
+    "eval_steps_per_second": 13.943
+}

logs/events.out.tfevents.1745278806.s_005_m.2836625.41

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3cd306fc151d43e60bb642b4cb5322b6e9fa1d9fa29b223f96d22794d3ff9bd5
+size 363

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 8.0,
+    "total_flos": 2.197388993748173e+16,
+    "train_loss": 0.4819710405861459,
+    "train_runtime": 453.5458,
+    "train_samples": 104743,
+    "train_samples_per_second": 11547.124,
+    "train_steps_per_second": 45.199
+}

trainer_state.json

@@ -0,0 +1,180 @@
+{
+  "best_global_step": 1230,
+  "best_metric": 0.6493877172470093,
+  "best_model_checkpoint": "tiny_bert_km_20_v2_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": 1.4939783811569214,
+      "learning_rate": 4.9e-05,
+      "loss": 0.6674,
+      "step": 410
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6101043382756727,
+      "eval_loss": 0.6505669951438904,
+      "eval_runtime": 1.5942,
+      "eval_samples_per_second": 3426.757,
+      "eval_steps_per_second": 13.8,
+      "step": 410
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.4395766258239746,
+      "learning_rate": 4.8e-05,
+      "loss": 0.6407,
+      "step": 820
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6229178107267069,
+      "eval_loss": 0.6511477828025818,
+      "eval_runtime": 1.5588,
+      "eval_samples_per_second": 3504.55,
+      "eval_steps_per_second": 14.113,
+      "step": 820
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 2.3813982009887695,
+      "learning_rate": 4.7e-05,
+      "loss": 0.5998,
+      "step": 1230
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6278601501006773,
+      "eval_loss": 0.6493877172470093,
+      "eval_runtime": 1.5636,
+      "eval_samples_per_second": 3493.938,
+      "eval_steps_per_second": 14.07,
+      "step": 1230
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 3.805351495742798,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.5319,
+      "step": 1640
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6154127768625297,
+      "eval_loss": 0.713868260383606,
+      "eval_runtime": 1.5488,
+      "eval_samples_per_second": 3527.196,
+      "eval_steps_per_second": 14.204,
+      "step": 1640
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 4.113601207733154,
+      "learning_rate": 4.5e-05,
+      "loss": 0.4555,
+      "step": 2050
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6117517847336629,
+      "eval_loss": 0.7634970545768738,
+      "eval_runtime": 1.5627,
+      "eval_samples_per_second": 3495.976,
+      "eval_steps_per_second": 14.079,
+      "step": 2050
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 4.7467193603515625,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.3843,
+      "step": 2460
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6071755445725792,
+      "eval_loss": 0.9000422954559326,
+      "eval_runtime": 1.5653,
+      "eval_samples_per_second": 3490.163,
+      "eval_steps_per_second": 14.055,
+      "step": 2460
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 5.9497246742248535,
+      "learning_rate": 4.3e-05,
+      "loss": 0.3163,
+      "step": 2870
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.604978949295259,
+      "eval_loss": 1.075139045715332,
+      "eval_runtime": 1.5751,
+      "eval_samples_per_second": 3468.419,
+      "eval_steps_per_second": 13.968,
+      "step": 2870
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 4.651660442352295,
+      "learning_rate": 4.2e-05,
+      "loss": 0.2599,
+      "step": 3280
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6077246933919092,
+      "eval_loss": 1.129774808883667,
+      "eval_runtime": 1.5511,
+      "eval_samples_per_second": 3521.95,
+      "eval_steps_per_second": 14.183,
+      "step": 3280
+    },
+    {
+      "epoch": 8.0,
+      "step": 3280,
+      "total_flos": 2.197388993748173e+16,
+      "train_loss": 0.4819710405861459,
+      "train_runtime": 453.5458,
+      "train_samples_per_second": 11547.124,
+      "train_steps_per_second": 45.199
+    }
+  ],
+  "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
+}