End of training

README.md

@@ -1,15 +1,33 @@
 ---
 library_name: transformers
+language:
+- en
 license: apache-2.0
 base_model: google/bert_uncased_L-4_H-512_A-8
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 - f1
 model-index:
 - name: bert_uncased_L-4_H-512_A-8_qqp
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE QQP
+      type: glue
+      args: qqp
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.8914172644076181
+    - name: F1
+      type: f1
+      value: 0.8539295933985492
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -17,12 +35,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # bert_uncased_L-4_H-512_A-8_qqp
 
-This model is a fine-tuned version of [google/bert_uncased_L-4_H-512_A-8](https://huggingface.co/google/bert_uncased_L-4_H-512_A-8) on an unknown dataset.
+This model is a fine-tuned version of [google/bert_uncased_L-4_H-512_A-8](https://huggingface.co/google/bert_uncased_L-4_H-512_A-8) on the GLUE QQP dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3568
-- Accuracy: 0.8931
-- F1: 0.8604
-- Combined Score: 0.8768
+- Loss: 0.2574
+- Accuracy: 0.8914
+- F1: 0.8539
+- Combined Score: 0.8727
 
 ## Model description
 

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.8914172644076181,
+    "eval_combined_score": 0.8726734289030837,
+    "eval_f1": 0.8539295933985492,
+    "eval_loss": 0.25741004943847656,
+    "eval_runtime": 15.8992,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 2542.902,
+    "eval_steps_per_second": 9.938,
+    "total_flos": 5.755995899574682e+16,
+    "train_loss": 0.18514271359235762,
+    "train_runtime": 1915.5632,
+    "train_samples": 363846,
+    "train_samples_per_second": 9497.103,
+    "train_steps_per_second": 37.117
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.8914172644076181,
+    "eval_combined_score": 0.8726734289030837,
+    "eval_f1": 0.8539295933985492,
+    "eval_loss": 0.25741004943847656,
+    "eval_runtime": 15.8992,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 2542.902,
+    "eval_steps_per_second": 9.938
+}

logs/events.out.tfevents.1733337689.ki-g0008.1761130.25

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3fdc8284a8277228121285f0e1783a114460688a4e7f1210102b5be0f705a79f
+size 515

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 8.0,
+    "total_flos": 5.755995899574682e+16,
+    "train_loss": 0.18514271359235762,
+    "train_runtime": 1915.5632,
+    "train_samples": 363846,
+    "train_samples_per_second": 9497.103,
+    "train_steps_per_second": 37.117
+}

trainer_state.json

@@ -0,0 +1,195 @@
+{
+  "best_metric": 0.25741004943847656,
+  "best_model_checkpoint": "bert_uncased_L-4_H-512_A-8_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": 3.9684581756591797,
+      "learning_rate": 4.9e-05,
+      "loss": 0.3551,
+      "step": 1422
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.86928023744744,
+      "eval_combined_score": 0.8435916399310384,
+      "eval_f1": 0.8179030424146366,
+      "eval_loss": 0.2993009090423584,
+      "eval_runtime": 16.3511,
+      "eval_samples_per_second": 2472.617,
+      "eval_steps_per_second": 9.663,
+      "step": 1422
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 4.823949813842773,
+      "learning_rate": 4.8e-05,
+      "loss": 0.2662,
+      "step": 2844
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8841701706653475,
+      "eval_combined_score": 0.8667011633726596,
+      "eval_f1": 0.8492321560799717,
+      "eval_loss": 0.2662528157234192,
+      "eval_runtime": 16.3577,
+      "eval_samples_per_second": 2471.626,
+      "eval_steps_per_second": 9.659,
+      "step": 2844
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 3.9237613677978516,
+      "learning_rate": 4.7e-05,
+      "loss": 0.217,
+      "step": 4266
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.8914172644076181,
+      "eval_combined_score": 0.8726734289030837,
+      "eval_f1": 0.8539295933985492,
+      "eval_loss": 0.25741004943847656,
+      "eval_runtime": 15.9668,
+      "eval_samples_per_second": 2532.133,
+      "eval_steps_per_second": 9.896,
+      "step": 4266
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 4.0418925285339355,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.179,
+      "step": 5688
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8936928023744743,
+      "eval_combined_score": 0.8743674804452473,
+      "eval_f1": 0.8550421585160203,
+      "eval_loss": 0.26455405354499817,
+      "eval_runtime": 16.0198,
+      "eval_samples_per_second": 2523.748,
+      "eval_steps_per_second": 9.863,
+      "step": 5688
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 3.7040185928344727,
+      "learning_rate": 4.5e-05,
+      "loss": 0.1487,
+      "step": 7110
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.8937917388078159,
+      "eval_combined_score": 0.8760050287514614,
+      "eval_f1": 0.8582183186951067,
+      "eval_loss": 0.291958212852478,
+      "eval_runtime": 16.0229,
+      "eval_samples_per_second": 2523.268,
+      "eval_steps_per_second": 9.861,
+      "step": 7110
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 4.520602226257324,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.1228,
+      "step": 8532
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.8936433341578036,
+      "eval_combined_score": 0.8761816355373063,
+      "eval_f1": 0.858719936916809,
+      "eval_loss": 0.29707542061805725,
+      "eval_runtime": 16.029,
+      "eval_samples_per_second": 2522.304,
+      "eval_steps_per_second": 9.857,
+      "step": 8532
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 3.891303062438965,
+      "learning_rate": 4.3e-05,
+      "loss": 0.1042,
+      "step": 9954
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.8915904031659658,
+      "eval_combined_score": 0.8746264294594484,
+      "eval_f1": 0.8576624557529309,
+      "eval_loss": 0.3390096127986908,
+      "eval_runtime": 16.0668,
+      "eval_samples_per_second": 2516.367,
+      "eval_steps_per_second": 9.834,
+      "step": 9954
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 5.342499256134033,
+      "learning_rate": 4.2e-05,
+      "loss": 0.0882,
+      "step": 11376
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.8930744496660895,
+      "eval_combined_score": 0.8767506365141864,
+      "eval_f1": 0.8604268233622833,
+      "eval_loss": 0.3567749857902527,
+      "eval_runtime": 16.0666,
+      "eval_samples_per_second": 2516.407,
+      "eval_steps_per_second": 9.834,
+      "step": 11376
+    },
+    {
+      "epoch": 8.0,
+      "step": 11376,
+      "total_flos": 5.755995899574682e+16,
+      "train_loss": 0.18514271359235762,
+      "train_runtime": 1915.5632,
+      "train_samples_per_second": 9497.103,
+      "train_steps_per_second": 37.117
+    }
+  ],
+  "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": 5.755995899574682e+16,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}