End of training

README.md

@@ -1,14 +1,32 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/bert_base_rand_10_v2
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 - f1
 model-index:
 - name: bert_base_rand_10_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: 0.8202572347266881
+    - name: F1
+      type: f1
+      value: 0.7589318294907945
 ---
 
 <!-- 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_v2_qqp
 
-This model is a fine-tuned version of [Hartunka/bert_base_rand_10_v2](https://huggingface.co/Hartunka/bert_base_rand_10_v2) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/bert_base_rand_10_v2](https://huggingface.co/Hartunka/bert_base_rand_10_v2) on the GLUE QQP dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6163
-- Accuracy: 0.8422
-- F1: 0.7848
-- Combined Score: 0.8135
+- Loss: 0.3871
+- Accuracy: 0.8203
+- F1: 0.7589
+- Combined Score: 0.7896
 
 ## Model description
 

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.8202572347266881,
+    "eval_combined_score": 0.7895945321087413,
+    "eval_f1": 0.7589318294907945,
+    "eval_loss": 0.3870999813079834,
+    "eval_runtime": 25.4254,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 1590.14,
+    "eval_steps_per_second": 6.214,
+    "total_flos": 3.3506166766980096e+17,
+    "train_loss": 0.26121269902096517,
+    "train_runtime": 4933.5908,
+    "train_samples": 363846,
+    "train_samples_per_second": 3687.436,
+    "train_steps_per_second": 14.411
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.8202572347266881,
+    "eval_combined_score": 0.7895945321087413,
+    "eval_f1": 0.7589318294907945,
+    "eval_loss": 0.3870999813079834,
+    "eval_runtime": 25.4254,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 1590.14,
+    "eval_steps_per_second": 6.214
+}

logs/events.out.tfevents.1745475842.s_005_m.2892304.43

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da58907dd40f5e9ffd2eed97a860b2398fd7209c91023cdb5066feb90bf4f875
+size 467

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 7.0,
+    "total_flos": 3.3506166766980096e+17,
+    "train_loss": 0.26121269902096517,
+    "train_runtime": 4933.5908,
+    "train_samples": 363846,
+    "train_samples_per_second": 3687.436,
+    "train_steps_per_second": 14.411
+}

trainer_state.json

@@ -0,0 +1,178 @@
+{
+  "best_global_step": 2844,
+  "best_metric": 0.3870999813079834,
+  "best_model_checkpoint": "bert_base_rand_10_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": 1.6563606262207031,
+      "learning_rate": 4.9e-05,
+      "loss": 0.4757,
+      "step": 1422
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7912935938659411,
+      "eval_combined_score": 0.7334524251045836,
+      "eval_f1": 0.6756112563432262,
+      "eval_loss": 0.4355120360851288,
+      "eval_runtime": 26.1103,
+      "eval_samples_per_second": 1548.43,
+      "eval_steps_per_second": 6.051,
+      "step": 1422
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 2.536411762237549,
+      "learning_rate": 4.8e-05,
+      "loss": 0.3701,
+      "step": 2844
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8202572347266881,
+      "eval_combined_score": 0.7895945321087413,
+      "eval_f1": 0.7589318294907945,
+      "eval_loss": 0.3870999813079834,
+      "eval_runtime": 25.5529,
+      "eval_samples_per_second": 1582.208,
+      "eval_steps_per_second": 6.183,
+      "step": 2844
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 2.4358298778533936,
+      "learning_rate": 4.7e-05,
+      "loss": 0.294,
+      "step": 4266
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.8242394261686866,
+      "eval_combined_score": 0.7994691106747047,
+      "eval_f1": 0.7746987951807229,
+      "eval_loss": 0.395740270614624,
+      "eval_runtime": 25.5956,
+      "eval_samples_per_second": 1579.568,
+      "eval_steps_per_second": 6.173,
+      "step": 4266
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 4.8816819190979,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.2331,
+      "step": 5688
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8342814741528568,
+      "eval_combined_score": 0.8015756756717092,
+      "eval_f1": 0.7688698771905617,
+      "eval_loss": 0.44763633608818054,
+      "eval_runtime": 25.5448,
+      "eval_samples_per_second": 1582.709,
+      "eval_steps_per_second": 6.185,
+      "step": 5688
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 2.7720863819122314,
+      "learning_rate": 4.5e-05,
+      "loss": 0.1845,
+      "step": 7110
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.8396240415533021,
+      "eval_combined_score": 0.8097644987128356,
+      "eval_f1": 0.7799049558723693,
+      "eval_loss": 0.4730183780193329,
+      "eval_runtime": 25.568,
+      "eval_samples_per_second": 1581.274,
+      "eval_steps_per_second": 6.18,
+      "step": 7110
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 4.084955215454102,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.1496,
+      "step": 8532
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.8420974523868414,
+      "eval_combined_score": 0.8117710815511749,
+      "eval_f1": 0.7814447107155084,
+      "eval_loss": 0.4950394928455353,
+      "eval_runtime": 25.5238,
+      "eval_samples_per_second": 1584.011,
+      "eval_steps_per_second": 6.19,
+      "step": 8532
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 6.653200149536133,
+      "learning_rate": 4.3e-05,
+      "loss": 0.1215,
+      "step": 9954
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.8421716547118476,
+      "eval_combined_score": 0.8134695424029774,
+      "eval_f1": 0.7847674300941073,
+      "eval_loss": 0.6163195967674255,
+      "eval_runtime": 25.5453,
+      "eval_samples_per_second": 1582.679,
+      "eval_steps_per_second": 6.185,
+      "step": 9954
+    },
+    {
+      "epoch": 7.0,
+      "step": 9954,
+      "total_flos": 3.3506166766980096e+17,
+      "train_loss": 0.26121269902096517,
+      "train_runtime": 4933.5908,
+      "train_samples_per_second": 3687.436,
+      "train_steps_per_second": 14.411
+    }
+  ],
+  "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
+}