End of training

README.md

@@ -1,14 +1,32 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/bert_base_rand_20_v2
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 - f1
 model-index:
 - name: bert_base_rand_20_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: 0.826020281968835
+    - name: F1
+      type: f1
+      value: 0.7738118206958647
 ---
 
 <!-- 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_20_v2_qqp
 
-This model is a fine-tuned version of [Hartunka/bert_base_rand_20_v2](https://huggingface.co/Hartunka/bert_base_rand_20_v2) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/bert_base_rand_20_v2](https://huggingface.co/Hartunka/bert_base_rand_20_v2) on the GLUE QQP dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5349
-- Accuracy: 0.8401
-- F1: 0.7878
-- Combined Score: 0.8139
+- Loss: 0.3937
+- Accuracy: 0.8260
+- F1: 0.7738
+- Combined Score: 0.7999
 
 ## Model description
 

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.826020281968835,
+    "eval_combined_score": 0.7999160513323499,
+    "eval_f1": 0.7738118206958647,
+    "eval_loss": 0.3936866819858551,
+    "eval_runtime": 25.4162,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 1590.717,
+    "eval_steps_per_second": 6.217,
+    "total_flos": 3.8292762019405824e+17,
+    "train_loss": 0.23944031840135277,
+    "train_runtime": 4918.4747,
+    "train_samples": 363846,
+    "train_samples_per_second": 3698.769,
+    "train_steps_per_second": 14.456
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.826020281968835,
+    "eval_combined_score": 0.7999160513323499,
+    "eval_f1": 0.7738118206958647,
+    "eval_loss": 0.3936866819858551,
+    "eval_runtime": 25.4162,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 1590.717,
+    "eval_steps_per_second": 6.217
+}

logs/events.out.tfevents.1745489920.s_005_m.2892304.61

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ece7e890e2c06fab21dcdffd3e905c1c2be7cc698c99cac7bf50f59d8aedf951
+size 467

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 8.0,
+    "total_flos": 3.8292762019405824e+17,
+    "train_loss": 0.23944031840135277,
+    "train_runtime": 4918.4747,
+    "train_samples": 363846,
+    "train_samples_per_second": 3698.769,
+    "train_steps_per_second": 14.456
+}

trainer_state.json

@@ -0,0 +1,196 @@
+{
+  "best_global_step": 4266,
+  "best_metric": 0.3936866819858551,
+  "best_model_checkpoint": "bert_base_rand_20_v2_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": 1.3981877565383911,
+      "learning_rate": 4.9e-05,
+      "loss": 0.4751,
+      "step": 1422
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7921840217660153,
+      "eval_combined_score": 0.7348128339788307,
+      "eval_f1": 0.6774416461916462,
+      "eval_loss": 0.4373413026332855,
+      "eval_runtime": 25.7712,
+      "eval_samples_per_second": 1568.804,
+      "eval_steps_per_second": 6.131,
+      "step": 1422
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 2.7053115367889404,
+      "learning_rate": 4.8e-05,
+      "loss": 0.3713,
+      "step": 2844
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8183032401681919,
+      "eval_combined_score": 0.7861771992993096,
+      "eval_f1": 0.7540511584304272,
+      "eval_loss": 0.39544248580932617,
+      "eval_runtime": 25.6338,
+      "eval_samples_per_second": 1577.213,
+      "eval_steps_per_second": 6.164,
+      "step": 2844
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 2.9879274368286133,
+      "learning_rate": 4.7e-05,
+      "loss": 0.2943,
+      "step": 4266
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.826020281968835,
+      "eval_combined_score": 0.7999160513323499,
+      "eval_f1": 0.7738118206958647,
+      "eval_loss": 0.3936866819858551,
+      "eval_runtime": 25.3325,
+      "eval_samples_per_second": 1595.973,
+      "eval_steps_per_second": 6.237,
+      "step": 4266
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 3.7611582279205322,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.2317,
+      "step": 5688
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8365322780113776,
+      "eval_combined_score": 0.8054576181986701,
+      "eval_f1": 0.7743829583859625,
+      "eval_loss": 0.43492114543914795,
+      "eval_runtime": 25.5862,
+      "eval_samples_per_second": 1580.146,
+      "eval_steps_per_second": 6.175,
+      "step": 5688
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 2.76202654838562,
+      "learning_rate": 4.5e-05,
+      "loss": 0.1827,
+      "step": 7110
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.839500371011625,
+      "eval_combined_score": 0.8076660317074914,
+      "eval_f1": 0.7758316924033578,
+      "eval_loss": 0.45615676045417786,
+      "eval_runtime": 25.5916,
+      "eval_samples_per_second": 1579.817,
+      "eval_steps_per_second": 6.174,
+      "step": 7110
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 3.0478901863098145,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.1456,
+      "step": 8532
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.8400445213950037,
+      "eval_combined_score": 0.8091454647574408,
+      "eval_f1": 0.7782464081198779,
+      "eval_loss": 0.5414330363273621,
+      "eval_runtime": 25.5831,
+      "eval_samples_per_second": 1580.339,
+      "eval_steps_per_second": 6.176,
+      "step": 8532
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 3.6382198333740234,
+      "learning_rate": 4.3e-05,
+      "loss": 0.1186,
+      "step": 9954
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.84232005936186,
+      "eval_combined_score": 0.813746381829203,
+      "eval_f1": 0.785172704296546,
+      "eval_loss": 0.6398206353187561,
+      "eval_runtime": 25.513,
+      "eval_samples_per_second": 1584.682,
+      "eval_steps_per_second": 6.193,
+      "step": 9954
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 6.167603969573975,
+      "learning_rate": 4.2e-05,
+      "loss": 0.0962,
+      "step": 11376
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.8400692555033391,
+      "eval_combined_score": 0.8139441900014562,
+      "eval_f1": 0.7878191244995734,
+      "eval_loss": 0.5349087119102478,
+      "eval_runtime": 25.5713,
+      "eval_samples_per_second": 1581.072,
+      "eval_steps_per_second": 6.179,
+      "step": 11376
+    },
+    {
+      "epoch": 8.0,
+      "step": 11376,
+      "total_flos": 3.8292762019405824e+17,
+      "train_loss": 0.23944031840135277,
+      "train_runtime": 4918.4747,
+      "train_samples_per_second": 3698.769,
+      "train_steps_per_second": 14.456
+    }
+  ],
+  "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
+}