End of training

README.md

@@ -1,13 +1,28 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/bert_base_km_10_v1
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 model-index:
 - name: bert_base_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: 0.6490939044481054
 ---
 
 <!-- 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. -->
 
 # bert_base_km_10_v1_qnli
 
-This model is a fine-tuned version of [Hartunka/bert_base_km_10_v1](https://huggingface.co/Hartunka/bert_base_km_10_v1) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/bert_base_km_10_v1](https://huggingface.co/Hartunka/bert_base_km_10_v1) on the GLUE QNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.2179
-- Accuracy: 0.6390
+- Loss: 0.6261
+- Accuracy: 0.6491
 
 ## Model description
 

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.6490939044481054,
+    "eval_loss": 0.6261129379272461,
+    "eval_runtime": 3.4175,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 1598.526,
+    "eval_steps_per_second": 6.437,
+    "total_flos": 9.645664445050368e+16,
+    "train_loss": 0.4108444413241609,
+    "train_runtime": 1229.6341,
+    "train_samples": 104743,
+    "train_samples_per_second": 4259.113,
+    "train_steps_per_second": 16.672
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.6490939044481054,
+    "eval_loss": 0.6261129379272461,
+    "eval_runtime": 3.4175,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 1598.526,
+    "eval_steps_per_second": 6.437
+}

logs/events.out.tfevents.1744978955.s_005_m.2772371.23

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:30f88c399bf21a8a765cd46a80220422ad9663fdf7edff7bee740d57bd7a4627
+size 363

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 7.0,
+    "total_flos": 9.645664445050368e+16,
+    "train_loss": 0.4108444413241609,
+    "train_runtime": 1229.6341,
+    "train_samples": 104743,
+    "train_samples_per_second": 4259.113,
+    "train_steps_per_second": 16.672
+}

trainer_state.json

@@ -0,0 +1,164 @@
+{
+  "best_global_step": 820,
+  "best_metric": 0.6261129379272461,
+  "best_model_checkpoint": "bert_base_km_10_v1_qnli/checkpoint-820",
+  "epoch": 7.0,
+  "eval_steps": 500,
+  "global_step": 2870,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 2.410963773727417,
+      "learning_rate": 4.9e-05,
+      "loss": 0.6641,
+      "step": 410
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6377448288486179,
+      "eval_loss": 0.6366901397705078,
+      "eval_runtime": 3.557,
+      "eval_samples_per_second": 1535.854,
+      "eval_steps_per_second": 6.185,
+      "step": 410
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 2.487971067428589,
+      "learning_rate": 4.8e-05,
+      "loss": 0.6198,
+      "step": 820
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6490939044481054,
+      "eval_loss": 0.6261129379272461,
+      "eval_runtime": 3.453,
+      "eval_samples_per_second": 1582.081,
+      "eval_steps_per_second": 6.371,
+      "step": 820
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 3.140673875808716,
+      "learning_rate": 4.7e-05,
+      "loss": 0.5408,
+      "step": 1230
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6265788028555739,
+      "eval_loss": 0.6862198114395142,
+      "eval_runtime": 3.4467,
+      "eval_samples_per_second": 1585.001,
+      "eval_steps_per_second": 6.383,
+      "step": 1230
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 7.148397922515869,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.4205,
+      "step": 1640
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6441515650741351,
+      "eval_loss": 0.7419140338897705,
+      "eval_runtime": 3.4555,
+      "eval_samples_per_second": 1580.954,
+      "eval_steps_per_second": 6.367,
+      "step": 1640
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 6.679465293884277,
+      "learning_rate": 4.5e-05,
+      "loss": 0.2925,
+      "step": 2050
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6448837634999085,
+      "eval_loss": 0.8849136233329773,
+      "eval_runtime": 3.4598,
+      "eval_samples_per_second": 1578.979,
+      "eval_steps_per_second": 6.359,
+      "step": 2050
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 8.848244667053223,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.1969,
+      "step": 2460
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.638843126487278,
+      "eval_loss": 1.09120774269104,
+      "eval_runtime": 3.4458,
+      "eval_samples_per_second": 1585.388,
+      "eval_steps_per_second": 6.385,
+      "step": 2460
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 9.282306671142578,
+      "learning_rate": 4.3e-05,
+      "loss": 0.1413,
+      "step": 2870
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6390261760937214,
+      "eval_loss": 1.2178927659988403,
+      "eval_runtime": 3.4453,
+      "eval_samples_per_second": 1585.62,
+      "eval_steps_per_second": 6.385,
+      "step": 2870
+    },
+    {
+      "epoch": 7.0,
+      "step": 2870,
+      "total_flos": 9.645664445050368e+16,
+      "train_loss": 0.4108444413241609,
+      "train_runtime": 1229.6341,
+      "train_samples_per_second": 4259.113,
+      "train_steps_per_second": 16.672
+    }
+  ],
+  "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": 9.645664445050368e+16,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}