End of training

README.md

@@ -1,14 +1,32 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/bert_base_km_5_v2
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 - f1
 model-index:
 - name: bert_base_km_5_v2_qqp
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE QQP
+      type: glue
+      args: qqp
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.8237447440019787
+    - name: F1
+      type: f1
+      value: 0.762450830055337
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -16,12 +34,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # bert_base_km_5_v2_qqp
 
-This model is a fine-tuned version of [Hartunka/bert_base_km_5_v2](https://huggingface.co/Hartunka/bert_base_km_5_v2) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/bert_base_km_5_v2](https://huggingface.co/Hartunka/bert_base_km_5_v2) on the GLUE QQP dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5664
-- Accuracy: 0.8406
-- F1: 0.7861
-- Combined Score: 0.8134
+- Loss: 0.3830
+- Accuracy: 0.8237
+- F1: 0.7625
+- Combined Score: 0.7931
 
 ## Model description
 

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.8237447440019787,
+    "eval_combined_score": 0.7930977870286579,
+    "eval_f1": 0.762450830055337,
+    "eval_loss": 0.38297006487846375,
+    "eval_runtime": 25.2546,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 1600.894,
+    "eval_steps_per_second": 6.256,
+    "total_flos": 3.3506166766980096e+17,
+    "train_loss": 0.23533505897146403,
+    "train_runtime": 4256.8509,
+    "train_samples": 363846,
+    "train_samples_per_second": 4273.652,
+    "train_steps_per_second": 16.702
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.8237447440019787,
+    "eval_combined_score": 0.7930977870286579,
+    "eval_f1": 0.762450830055337,
+    "eval_loss": 0.38297006487846375,
+    "eval_runtime": 25.2546,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 1600.894,
+    "eval_steps_per_second": 6.256
+}

logs/events.out.tfevents.1745273967.s_005_m.2836894.7

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6165e5ebca01eae854ea0e31f4049df8a4d5e4861b84286a640624055f3f946d
+size 467

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 7.0,
+    "total_flos": 3.3506166766980096e+17,
+    "train_loss": 0.23533505897146403,
+    "train_runtime": 4256.8509,
+    "train_samples": 363846,
+    "train_samples_per_second": 4273.652,
+    "train_steps_per_second": 16.702
+}

trainer_state.json

@@ -0,0 +1,178 @@
+{
+  "best_global_step": 2844,
+  "best_metric": 0.38297006487846375,
+  "best_model_checkpoint": "bert_base_km_5_v2_qqp/checkpoint-2844",
+  "epoch": 7.0,
+  "eval_steps": 500,
+  "global_step": 9954,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 2.0869553089141846,
+      "learning_rate": 4.9e-05,
+      "loss": 0.4655,
+      "step": 1422
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7970813752164234,
+      "eval_combined_score": 0.7420713559023833,
+      "eval_f1": 0.687061336588343,
+      "eval_loss": 0.42958006262779236,
+      "eval_runtime": 25.6003,
+      "eval_samples_per_second": 1579.279,
+      "eval_steps_per_second": 6.172,
+      "step": 1422
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 3.090320348739624,
+      "learning_rate": 4.8e-05,
+      "loss": 0.3502,
+      "step": 2844
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8237447440019787,
+      "eval_combined_score": 0.7930977870286579,
+      "eval_f1": 0.762450830055337,
+      "eval_loss": 0.38297006487846375,
+      "eval_runtime": 25.307,
+      "eval_samples_per_second": 1597.58,
+      "eval_steps_per_second": 6.243,
+      "step": 2844
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 3.375284194946289,
+      "learning_rate": 4.7e-05,
+      "loss": 0.2673,
+      "step": 4266
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.8350234974029186,
+      "eval_combined_score": 0.8055012697601011,
+      "eval_f1": 0.7759790421172835,
+      "eval_loss": 0.4027860760688782,
+      "eval_runtime": 25.2732,
+      "eval_samples_per_second": 1599.72,
+      "eval_steps_per_second": 6.252,
+      "step": 4266
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 4.9200439453125,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.2003,
+      "step": 5688
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8395745733366312,
+      "eval_combined_score": 0.8054280229684778,
+      "eval_f1": 0.7712814726003244,
+      "eval_loss": 0.4558044970035553,
+      "eval_runtime": 25.3269,
+      "eval_samples_per_second": 1596.328,
+      "eval_steps_per_second": 6.238,
+      "step": 5688
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 3.585550546646118,
+      "learning_rate": 4.5e-05,
+      "loss": 0.151,
+      "step": 7110
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.843705169428642,
+      "eval_combined_score": 0.8116771828243046,
+      "eval_f1": 0.7796491962199672,
+      "eval_loss": 0.4538190960884094,
+      "eval_runtime": 25.1996,
+      "eval_samples_per_second": 1604.391,
+      "eval_steps_per_second": 6.27,
+      "step": 7110
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 3.546767234802246,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.1178,
+      "step": 8532
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.8423942616868662,
+      "eval_combined_score": 0.8112895828925568,
+      "eval_f1": 0.7801849040982476,
+      "eval_loss": 0.556125819683075,
+      "eval_runtime": 25.1928,
+      "eval_samples_per_second": 1604.822,
+      "eval_steps_per_second": 6.272,
+      "step": 8532
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 4.131615161895752,
+      "learning_rate": 4.3e-05,
+      "loss": 0.0952,
+      "step": 9954
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.8405886717783824,
+      "eval_combined_score": 0.8133517812149138,
+      "eval_f1": 0.7861148906514452,
+      "eval_loss": 0.5664165616035461,
+      "eval_runtime": 25.2685,
+      "eval_samples_per_second": 1600.015,
+      "eval_steps_per_second": 6.253,
+      "step": 9954
+    },
+    {
+      "epoch": 7.0,
+      "step": 9954,
+      "total_flos": 3.3506166766980096e+17,
+      "train_loss": 0.23533505897146403,
+      "train_runtime": 4256.8509,
+      "train_samples_per_second": 4273.652,
+      "train_steps_per_second": 16.702
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 71100,
+  "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": 3.3506166766980096e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}