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-256_A-4
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 - f1
 model-index:
 - name: bert_uncased_L-4_H-256_A-4_qqp
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE QQP
+      type: glue
+      args: qqp
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.8774672273064557
+    - name: F1
+      type: f1
+      value: 0.8326577489528443
 ---
 
 <!-- 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-256_A-4_qqp
 
-This model is a fine-tuned version of [google/bert_uncased_L-4_H-256_A-4](https://huggingface.co/google/bert_uncased_L-4_H-256_A-4) on an unknown dataset.
+This model is a fine-tuned version of [google/bert_uncased_L-4_H-256_A-4](https://huggingface.co/google/bert_uncased_L-4_H-256_A-4) on the GLUE QQP dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3125
-- Accuracy: 0.8863
-- F1: 0.8526
-- Combined Score: 0.8694
+- Loss: 0.2840
+- Accuracy: 0.8775
+- F1: 0.8327
+- Combined Score: 0.8551
 
 ## Model description
 

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 9.0,
+    "eval_accuracy": 0.8774672273064557,
+    "eval_combined_score": 0.85506248812965,
+    "eval_f1": 0.8326577489528443,
+    "eval_loss": 0.28400716185569763,
+    "eval_runtime": 14.2109,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 2845.0,
+    "eval_steps_per_second": 11.118,
+    "total_flos": 1.6225443484895232e+16,
+    "train_loss": 0.2465917483253914,
+    "train_runtime": 1672.4569,
+    "train_samples": 363846,
+    "train_samples_per_second": 10877.59,
+    "train_steps_per_second": 42.512
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 9.0,
+    "eval_accuracy": 0.8774672273064557,
+    "eval_combined_score": 0.85506248812965,
+    "eval_f1": 0.8326577489528443,
+    "eval_loss": 0.28400716185569763,
+    "eval_runtime": 14.2109,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 2845.0,
+    "eval_steps_per_second": 11.118
+}

logs/events.out.tfevents.1733332632.ki-g0008.1761130.7

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01015c9dc853728c03d525ea0161064afebbc6ff0f96dda12c332e3b7d3dfbeb
+size 515

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 9.0,
+    "total_flos": 1.6225443484895232e+16,
+    "train_loss": 0.2465917483253914,
+    "train_runtime": 1672.4569,
+    "train_samples": 363846,
+    "train_samples_per_second": 10877.59,
+    "train_steps_per_second": 42.512
+}

trainer_state.json

@@ -0,0 +1,213 @@
+{
+  "best_metric": 0.28400716185569763,
+  "best_model_checkpoint": "bert_uncased_L-4_H-256_A-4_qqp/checkpoint-5688",
+  "epoch": 9.0,
+  "eval_steps": 500,
+  "global_step": 12798,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 2.7269396781921387,
+      "learning_rate": 4.9e-05,
+      "loss": 0.3985,
+      "step": 1422
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.8486272569873856,
+      "eval_combined_score": 0.8226119534072797,
+      "eval_f1": 0.7965966498271737,
+      "eval_loss": 0.3341473639011383,
+      "eval_runtime": 14.849,
+      "eval_samples_per_second": 2722.744,
+      "eval_steps_per_second": 10.64,
+      "step": 1422
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 5.564885139465332,
+      "learning_rate": 4.8e-05,
+      "loss": 0.3199,
+      "step": 2844
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8635913925302993,
+      "eval_combined_score": 0.8440247442328105,
+      "eval_f1": 0.8244580959353217,
+      "eval_loss": 0.3058396279811859,
+      "eval_runtime": 14.2665,
+      "eval_samples_per_second": 2833.914,
+      "eval_steps_per_second": 11.075,
+      "step": 2844
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 3.7981770038604736,
+      "learning_rate": 4.7e-05,
+      "loss": 0.2819,
+      "step": 4266
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.8731634924560969,
+      "eval_combined_score": 0.8536419778383222,
+      "eval_f1": 0.8341204632205473,
+      "eval_loss": 0.28831592202186584,
+      "eval_runtime": 14.2607,
+      "eval_samples_per_second": 2835.056,
+      "eval_steps_per_second": 11.079,
+      "step": 4266
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 3.4819419384002686,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.2525,
+      "step": 5688
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8774672273064557,
+      "eval_combined_score": 0.85506248812965,
+      "eval_f1": 0.8326577489528443,
+      "eval_loss": 0.28400716185569763,
+      "eval_runtime": 15.0424,
+      "eval_samples_per_second": 2687.732,
+      "eval_steps_per_second": 10.504,
+      "step": 5688
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 5.758821964263916,
+      "learning_rate": 4.5e-05,
+      "loss": 0.2304,
+      "step": 7110
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.8807815978233985,
+      "eval_combined_score": 0.8627940444523007,
+      "eval_f1": 0.8448064910812029,
+      "eval_loss": 0.2858343720436096,
+      "eval_runtime": 15.1683,
+      "eval_samples_per_second": 2665.432,
+      "eval_steps_per_second": 10.416,
+      "step": 7110
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 3.458667039871216,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.2094,
+      "step": 8532
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.8816967598318081,
+      "eval_combined_score": 0.8633256077028664,
+      "eval_f1": 0.8449544555739247,
+      "eval_loss": 0.2876797616481781,
+      "eval_runtime": 14.2687,
+      "eval_samples_per_second": 2833.468,
+      "eval_steps_per_second": 11.073,
+      "step": 8532
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 5.1516289710998535,
+      "learning_rate": 4.3e-05,
+      "loss": 0.1912,
+      "step": 9954
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.8822903784318575,
+      "eval_combined_score": 0.8642357868435184,
+      "eval_f1": 0.8461811952551794,
+      "eval_loss": 0.29094812273979187,
+      "eval_runtime": 14.2811,
+      "eval_samples_per_second": 2831.012,
+      "eval_steps_per_second": 11.064,
+      "step": 9954
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 3.7632157802581787,
+      "learning_rate": 4.2e-05,
+      "loss": 0.1749,
+      "step": 11376
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.8856294830571357,
+      "eval_combined_score": 0.8683925436273253,
+      "eval_f1": 0.8511556041975149,
+      "eval_loss": 0.29442498087882996,
+      "eval_runtime": 14.2478,
+      "eval_samples_per_second": 2837.639,
+      "eval_steps_per_second": 11.089,
+      "step": 11376
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 5.199641227722168,
+      "learning_rate": 4.1e-05,
+      "loss": 0.1604,
+      "step": 12798
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.886272569873856,
+      "eval_combined_score": 0.8694173084731301,
+      "eval_f1": 0.8525620470724042,
+      "eval_loss": 0.3124821186065674,
+      "eval_runtime": 14.2362,
+      "eval_samples_per_second": 2839.944,
+      "eval_steps_per_second": 11.098,
+      "step": 12798
+    },
+    {
+      "epoch": 9.0,
+      "step": 12798,
+      "total_flos": 1.6225443484895232e+16,
+      "train_loss": 0.2465917483253914,
+      "train_runtime": 1672.4569,
+      "train_samples_per_second": 10877.59,
+      "train_steps_per_second": 42.512
+    }
+  ],
+  "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": 1.6225443484895232e+16,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}