End of training

README.md

@@ -1,14 +1,32 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/bert_base_rand_10_v1
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 - f1
 model-index:
 - name: bert_base_rand_10_v1_qqp
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE QQP
+      type: glue
+      args: qqp
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.82883997031907
+    - name: F1
+      type: f1
+      value: 0.7734119187950229
 ---
 
 <!-- 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_rand_10_v1_qqp
 
-This model is a fine-tuned version of [Hartunka/bert_base_rand_10_v1](https://huggingface.co/Hartunka/bert_base_rand_10_v1) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/bert_base_rand_10_v1](https://huggingface.co/Hartunka/bert_base_rand_10_v1) on the GLUE QQP dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5849
-- Accuracy: 0.8363
-- F1: 0.7872
-- Combined Score: 0.8117
+- Loss: 0.3889
+- Accuracy: 0.8288
+- F1: 0.7734
+- Combined Score: 0.8011
 
 ## Model description
 

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.82883997031907,
+    "eval_combined_score": 0.8011259445570464,
+    "eval_f1": 0.7734119187950229,
+    "eval_loss": 0.38887616991996765,
+    "eval_runtime": 25.0972,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 1610.938,
+    "eval_steps_per_second": 6.296,
+    "total_flos": 3.8292762019405824e+17,
+    "train_loss": 0.24114117575429495,
+    "train_runtime": 4872.8247,
+    "train_samples": 363846,
+    "train_samples_per_second": 3733.42,
+    "train_steps_per_second": 14.591
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.82883997031907,
+    "eval_combined_score": 0.8011259445570464,
+    "eval_f1": 0.7734119187950229,
+    "eval_loss": 0.38887616991996765,
+    "eval_runtime": 25.0972,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 1610.938,
+    "eval_steps_per_second": 6.296
+}

logs/events.out.tfevents.1745139733.s_005_m.2796520.61

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7e9b81238f175d965ea1981cd87914b1c1839dae23780b767e462e944b37a8fe
+size 467

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 8.0,
+    "total_flos": 3.8292762019405824e+17,
+    "train_loss": 0.24114117575429495,
+    "train_runtime": 4872.8247,
+    "train_samples": 363846,
+    "train_samples_per_second": 3733.42,
+    "train_steps_per_second": 14.591
+}

trainer_state.json

@@ -0,0 +1,196 @@
+{
+  "best_global_step": 4266,
+  "best_metric": 0.38887616991996765,
+  "best_model_checkpoint": "bert_base_rand_10_v1_qqp/checkpoint-4266",
+  "epoch": 8.0,
+  "eval_steps": 500,
+  "global_step": 11376,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.5187573432922363,
+      "learning_rate": 4.9e-05,
+      "loss": 0.4757,
+      "step": 1422
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7961909473163492,
+      "eval_combined_score": 0.7495413523630134,
+      "eval_f1": 0.7028917574096777,
+      "eval_loss": 0.42474642395973206,
+      "eval_runtime": 25.6784,
+      "eval_samples_per_second": 1574.473,
+      "eval_steps_per_second": 6.153,
+      "step": 1422
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 2.3227264881134033,
+      "learning_rate": 4.8e-05,
+      "loss": 0.3697,
+      "step": 2844
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8153351471679446,
+      "eval_combined_score": 0.7867880968661431,
+      "eval_f1": 0.7582410465643417,
+      "eval_loss": 0.3916879892349243,
+      "eval_runtime": 25.2045,
+      "eval_samples_per_second": 1604.077,
+      "eval_steps_per_second": 6.269,
+      "step": 2844
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 3.029405117034912,
+      "learning_rate": 4.7e-05,
+      "loss": 0.2939,
+      "step": 4266
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.82883997031907,
+      "eval_combined_score": 0.8011259445570464,
+      "eval_f1": 0.7734119187950229,
+      "eval_loss": 0.38887616991996765,
+      "eval_runtime": 25.2165,
+      "eval_samples_per_second": 1603.318,
+      "eval_steps_per_second": 6.266,
+      "step": 4266
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 4.556088447570801,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.2329,
+      "step": 5688
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8375958446697996,
+      "eval_combined_score": 0.8048760679383421,
+      "eval_f1": 0.7721562912068846,
+      "eval_loss": 0.4263524115085602,
+      "eval_runtime": 25.2524,
+      "eval_samples_per_second": 1601.037,
+      "eval_steps_per_second": 6.257,
+      "step": 5688
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 3.429358720779419,
+      "learning_rate": 4.5e-05,
+      "loss": 0.1847,
+      "step": 7110
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.8403660648033638,
+      "eval_combined_score": 0.8105553655991272,
+      "eval_f1": 0.7807446663948906,
+      "eval_loss": 0.45088598132133484,
+      "eval_runtime": 25.2221,
+      "eval_samples_per_second": 1602.959,
+      "eval_steps_per_second": 6.264,
+      "step": 7110
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 2.867098093032837,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.1492,
+      "step": 8532
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.8399455849616622,
+      "eval_combined_score": 0.8100152733644335,
+      "eval_f1": 0.7800849617672048,
+      "eval_loss": 0.4775959253311157,
+      "eval_runtime": 25.2785,
+      "eval_samples_per_second": 1599.381,
+      "eval_steps_per_second": 6.25,
+      "step": 8532
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 3.9507932662963867,
+      "learning_rate": 4.3e-05,
+      "loss": 0.122,
+      "step": 9954
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.8428642097452387,
+      "eval_combined_score": 0.8128746393041152,
+      "eval_f1": 0.7828850688629917,
+      "eval_loss": 0.5778086185455322,
+      "eval_runtime": 25.2512,
+      "eval_samples_per_second": 1601.114,
+      "eval_steps_per_second": 6.257,
+      "step": 9954
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 4.067262172698975,
+      "learning_rate": 4.2e-05,
+      "loss": 0.1009,
+      "step": 11376
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.8362602028196884,
+      "eval_combined_score": 0.8117266039172377,
+      "eval_f1": 0.7871930050147872,
+      "eval_loss": 0.5848562717437744,
+      "eval_runtime": 25.2208,
+      "eval_samples_per_second": 1603.042,
+      "eval_steps_per_second": 6.265,
+      "step": 11376
+    },
+    {
+      "epoch": 8.0,
+      "step": 11376,
+      "total_flos": 3.8292762019405824e+17,
+      "train_loss": 0.24114117575429495,
+      "train_runtime": 4872.8247,
+      "train_samples_per_second": 3733.42,
+      "train_steps_per_second": 14.591
+    }
+  ],
+  "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": 3.8292762019405824e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}