End of training

README.md

@@ -1,14 +1,32 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/bert_base_rand_20_v1
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 - f1
 model-index:
 - name: bert_base_rand_20_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: 0.8193420727182785
+    - name: F1
+      type: f1
+      value: 0.7553098827470687
 ---
 
 <!-- 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_v1_qqp
 
-This model is a fine-tuned version of [Hartunka/bert_base_rand_20_v1](https://huggingface.co/Hartunka/bert_base_rand_20_v1) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/bert_base_rand_20_v1](https://huggingface.co/Hartunka/bert_base_rand_20_v1) on the GLUE QQP dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5579
-- Accuracy: 0.8411
-- F1: 0.7886
-- Combined Score: 0.8149
+- Loss: 0.3930
+- Accuracy: 0.8193
+- F1: 0.7553
+- Combined Score: 0.7873
 
 ## Model description
 

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.8193420727182785,
+    "eval_combined_score": 0.7873259777326735,
+    "eval_f1": 0.7553098827470687,
+    "eval_loss": 0.3929571211338043,
+    "eval_runtime": 25.0712,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 1612.606,
+    "eval_steps_per_second": 6.302,
+    "total_flos": 3.3506166766980096e+17,
+    "train_loss": 0.25951667403944867,
+    "train_runtime": 4265.3847,
+    "train_samples": 363846,
+    "train_samples_per_second": 4265.102,
+    "train_steps_per_second": 16.669
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.8193420727182785,
+    "eval_combined_score": 0.7873259777326735,
+    "eval_f1": 0.7553098827470687,
+    "eval_loss": 0.3929571211338043,
+    "eval_runtime": 25.0712,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 1612.606,
+    "eval_steps_per_second": 6.302
+}

logs/events.out.tfevents.1745153114.s_005_m.2796520.79

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5ba4e5d293627c437dae33d2fcdff0093f218b8a7b59bd91ee851e2bac18f33
+size 467

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 7.0,
+    "total_flos": 3.3506166766980096e+17,
+    "train_loss": 0.25951667403944867,
+    "train_runtime": 4265.3847,
+    "train_samples": 363846,
+    "train_samples_per_second": 4265.102,
+    "train_steps_per_second": 16.669
+}

trainer_state.json

@@ -0,0 +1,178 @@
+{
+  "best_global_step": 2844,
+  "best_metric": 0.3929571211338043,
+  "best_model_checkpoint": "bert_base_rand_20_v1_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": 1.445243000984192,
+      "learning_rate": 4.9e-05,
+      "loss": 0.4749,
+      "step": 1422
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7926292357160525,
+      "eval_combined_score": 0.7384832925568214,
+      "eval_f1": 0.6843373493975904,
+      "eval_loss": 0.4352428615093231,
+      "eval_runtime": 25.4547,
+      "eval_samples_per_second": 1588.311,
+      "eval_steps_per_second": 6.207,
+      "step": 1422
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 2.5577476024627686,
+      "learning_rate": 4.8e-05,
+      "loss": 0.3711,
+      "step": 2844
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8193420727182785,
+      "eval_combined_score": 0.7873259777326735,
+      "eval_f1": 0.7553098827470687,
+      "eval_loss": 0.3929571211338043,
+      "eval_runtime": 25.2514,
+      "eval_samples_per_second": 1601.1,
+      "eval_steps_per_second": 6.257,
+      "step": 2844
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 2.8862438201904297,
+      "learning_rate": 4.7e-05,
+      "loss": 0.2943,
+      "step": 4266
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.8286173633440514,
+      "eval_combined_score": 0.8017993451345009,
+      "eval_f1": 0.7749813269249505,
+      "eval_loss": 0.39898237586021423,
+      "eval_runtime": 25.2275,
+      "eval_samples_per_second": 1602.617,
+      "eval_steps_per_second": 6.263,
+      "step": 4266
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 3.059464454650879,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.2315,
+      "step": 5688
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8354439772446204,
+      "eval_combined_score": 0.8015854107498805,
+      "eval_f1": 0.7677268442551408,
+      "eval_loss": 0.4510485827922821,
+      "eval_runtime": 25.2305,
+      "eval_samples_per_second": 1602.429,
+      "eval_steps_per_second": 6.262,
+      "step": 5688
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 2.1476404666900635,
+      "learning_rate": 4.5e-05,
+      "loss": 0.1825,
+      "step": 7110
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.8380905268365075,
+      "eval_combined_score": 0.8061053048261213,
+      "eval_f1": 0.774120082815735,
+      "eval_loss": 0.44589605927467346,
+      "eval_runtime": 25.2283,
+      "eval_samples_per_second": 1602.564,
+      "eval_steps_per_second": 6.263,
+      "step": 7110
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 3.517934799194336,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.1449,
+      "step": 8532
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.8404897353450408,
+      "eval_combined_score": 0.8084344720277339,
+      "eval_f1": 0.7763792087104269,
+      "eval_loss": 0.5557775497436523,
+      "eval_runtime": 25.2154,
+      "eval_samples_per_second": 1603.388,
+      "eval_steps_per_second": 6.266,
+      "step": 8532
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 4.514515399932861,
+      "learning_rate": 4.3e-05,
+      "loss": 0.1174,
+      "step": 9954
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.8411080880534256,
+      "eval_combined_score": 0.8148544257563429,
+      "eval_f1": 0.7886007634592602,
+      "eval_loss": 0.557855486869812,
+      "eval_runtime": 25.3627,
+      "eval_samples_per_second": 1594.076,
+      "eval_steps_per_second": 6.23,
+      "step": 9954
+    },
+    {
+      "epoch": 7.0,
+      "step": 9954,
+      "total_flos": 3.3506166766980096e+17,
+      "train_loss": 0.25951667403944867,
+      "train_runtime": 4265.3847,
+      "train_samples_per_second": 4265.102,
+      "train_steps_per_second": 16.669
+    }
+  ],
+  "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
+}