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-128_A-2
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 - f1
 model-index:
 - name: bert_uncased_L-4_H-128_A-2_mrpc
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE MRPC
+      type: glue
+      args: mrpc
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.7573529411764706
+    - name: F1
+      type: f1
+      value: 0.840064620355412
 ---
 
 <!-- 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-128_A-2_mrpc
 
-This model is a fine-tuned version of [google/bert_uncased_L-4_H-128_A-2](https://huggingface.co/google/bert_uncased_L-4_H-128_A-2) on an unknown dataset.
+This model is a fine-tuned version of [google/bert_uncased_L-4_H-128_A-2](https://huggingface.co/google/bert_uncased_L-4_H-128_A-2) on the GLUE MRPC dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5588
-- Accuracy: 0.7525
-- F1: 0.8374
-- Combined Score: 0.7949
+- Loss: 0.5327
+- Accuracy: 0.7574
+- F1: 0.8401
+- Combined Score: 0.7987
 
 ## Model description
 

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 16.0,
+    "eval_accuracy": 0.7573529411764706,
+    "eval_combined_score": 0.7987087807659412,
+    "eval_f1": 0.840064620355412,
+    "eval_loss": 0.5326957106590271,
+    "eval_runtime": 0.1586,
+    "eval_samples": 408,
+    "eval_samples_per_second": 2573.297,
+    "eval_steps_per_second": 12.614,
+    "total_flos": 73027538583552.0,
+    "train_loss": 0.490890504916509,
+    "train_runtime": 29.3983,
+    "train_samples": 3668,
+    "train_samples_per_second": 6238.45,
+    "train_steps_per_second": 25.512
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 16.0,
+    "eval_accuracy": 0.7573529411764706,
+    "eval_combined_score": 0.7987087807659412,
+    "eval_f1": 0.840064620355412,
+    "eval_loss": 0.5326957106590271,
+    "eval_runtime": 0.1586,
+    "eval_samples": 408,
+    "eval_samples_per_second": 2573.297,
+    "eval_steps_per_second": 12.614
+}

logs/events.out.tfevents.1733335350.ki-g0008.1764072.21

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6d9732eac57f5cd01615211bea80b46e465018de2041b1b7ea54ad713f95ca8a
+size 515

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 16.0,
+    "total_flos": 73027538583552.0,
+    "train_loss": 0.490890504916509,
+    "train_runtime": 29.3983,
+    "train_samples": 3668,
+    "train_samples_per_second": 6238.45,
+    "train_steps_per_second": 25.512
+}

trainer_state.json

@@ -0,0 +1,339 @@
+{
+  "best_metric": 0.5326957106590271,
+  "best_model_checkpoint": "bert_uncased_L-4_H-128_A-2_mrpc/checkpoint-165",
+  "epoch": 16.0,
+  "eval_steps": 500,
+  "global_step": 240,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.4740382730960846,
+      "learning_rate": 4.9e-05,
+      "loss": 0.6437,
+      "step": 15
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7480253018237863,
+      "eval_f1": 0.8122270742358079,
+      "eval_loss": 0.6180539727210999,
+      "eval_runtime": 0.1512,
+      "eval_samples_per_second": 2698.339,
+      "eval_steps_per_second": 13.227,
+      "step": 15
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.8819882273674011,
+      "learning_rate": 4.8e-05,
+      "loss": 0.6197,
+      "step": 30
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6911764705882353,
+      "eval_combined_score": 0.7534829721362228,
+      "eval_f1": 0.8157894736842105,
+      "eval_loss": 0.6046980023384094,
+      "eval_runtime": 0.1487,
+      "eval_samples_per_second": 2742.909,
+      "eval_steps_per_second": 13.446,
+      "step": 30
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 2.03017258644104,
+      "learning_rate": 4.7e-05,
+      "loss": 0.595,
+      "step": 45
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6985294117647058,
+      "eval_combined_score": 0.7573364547612766,
+      "eval_f1": 0.8161434977578476,
+      "eval_loss": 0.5876799821853638,
+      "eval_runtime": 0.1465,
+      "eval_samples_per_second": 2785.842,
+      "eval_steps_per_second": 13.656,
+      "step": 45
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.8701429963111877,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.582,
+      "step": 60
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7279411764705882,
+      "eval_combined_score": 0.7781900627329756,
+      "eval_f1": 0.8284389489953632,
+      "eval_loss": 0.5686801075935364,
+      "eval_runtime": 0.1522,
+      "eval_samples_per_second": 2680.994,
+      "eval_steps_per_second": 13.142,
+      "step": 60
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.5039589405059814,
+      "learning_rate": 4.5e-05,
+      "loss": 0.5617,
+      "step": 75
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.7279411764705882,
+      "eval_combined_score": 0.778717132014096,
+      "eval_f1": 0.8294930875576038,
+      "eval_loss": 0.5593908429145813,
+      "eval_runtime": 0.1469,
+      "eval_samples_per_second": 2777.387,
+      "eval_steps_per_second": 13.615,
+      "step": 75
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.3441897630691528,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.5409,
+      "step": 90
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.7132352941176471,
+      "eval_combined_score": 0.7670311233222232,
+      "eval_f1": 0.8208269525267994,
+      "eval_loss": 0.5550417304039001,
+      "eval_runtime": 0.1443,
+      "eval_samples_per_second": 2828.2,
+      "eval_steps_per_second": 13.864,
+      "step": 90
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 2.370797872543335,
+      "learning_rate": 4.3e-05,
+      "loss": 0.5213,
+      "step": 105
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.7254901960784313,
+      "eval_combined_score": 0.7749708033683693,
+      "eval_f1": 0.8244514106583073,
+      "eval_loss": 0.541745662689209,
+      "eval_runtime": 0.1475,
+      "eval_samples_per_second": 2766.445,
+      "eval_steps_per_second": 13.561,
+      "step": 105
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 3.4701988697052,
+      "learning_rate": 4.2e-05,
+      "loss": 0.4968,
+      "step": 120
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.7328431372549019,
+      "eval_combined_score": 0.7819254445964432,
+      "eval_f1": 0.8310077519379845,
+      "eval_loss": 0.5529593825340271,
+      "eval_runtime": 0.1516,
+      "eval_samples_per_second": 2690.55,
+      "eval_steps_per_second": 13.189,
+      "step": 120
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 2.4742236137390137,
+      "learning_rate": 4.1e-05,
+      "loss": 0.4741,
+      "step": 135
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.7352941176470589,
+      "eval_combined_score": 0.7843137254901962,
+      "eval_f1": 0.8333333333333334,
+      "eval_loss": 0.5579618811607361,
+      "eval_runtime": 0.1396,
+      "eval_samples_per_second": 2922.56,
+      "eval_steps_per_second": 14.326,
+      "step": 135
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 3.9524009227752686,
+      "learning_rate": 4e-05,
+      "loss": 0.4545,
+      "step": 150
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.7549019607843137,
+      "eval_combined_score": 0.7973227752639517,
+      "eval_f1": 0.8397435897435896,
+      "eval_loss": 0.539033830165863,
+      "eval_runtime": 0.1482,
+      "eval_samples_per_second": 2753.474,
+      "eval_steps_per_second": 13.497,
+      "step": 150
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 2.9584977626800537,
+      "learning_rate": 3.9000000000000006e-05,
+      "loss": 0.4366,
+      "step": 165
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.7573529411764706,
+      "eval_combined_score": 0.7987087807659412,
+      "eval_f1": 0.840064620355412,
+      "eval_loss": 0.5326957106590271,
+      "eval_runtime": 0.1409,
+      "eval_samples_per_second": 2895.937,
+      "eval_steps_per_second": 14.196,
+      "step": 165
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 5.0168890953063965,
+      "learning_rate": 3.8e-05,
+      "loss": 0.4206,
+      "step": 180
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.7598039215686274,
+      "eval_combined_score": 0.8011238257360822,
+      "eval_f1": 0.842443729903537,
+      "eval_loss": 0.535027027130127,
+      "eval_runtime": 0.1403,
+      "eval_samples_per_second": 2907.08,
+      "eval_steps_per_second": 14.25,
+      "step": 180
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 2.828728437423706,
+      "learning_rate": 3.7e-05,
+      "loss": 0.397,
+      "step": 195
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.7549019607843137,
+      "eval_combined_score": 0.7998112288393618,
+      "eval_f1": 0.84472049689441,
+      "eval_loss": 0.5649162530899048,
+      "eval_runtime": 0.1409,
+      "eval_samples_per_second": 2895.59,
+      "eval_steps_per_second": 14.194,
+      "step": 195
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 6.568640232086182,
+      "learning_rate": 3.6e-05,
+      "loss": 0.3873,
+      "step": 210
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.7622549019607843,
+      "eval_combined_score": 0.8052276074749147,
+      "eval_f1": 0.8482003129890453,
+      "eval_loss": 0.5601567029953003,
+      "eval_runtime": 0.1451,
+      "eval_samples_per_second": 2812.318,
+      "eval_steps_per_second": 13.786,
+      "step": 210
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 3.909018039703369,
+      "learning_rate": 3.5e-05,
+      "loss": 0.3725,
+      "step": 225
+    },
+    {
+      "epoch": 15.0,
+      "eval_accuracy": 0.7524509803921569,
+      "eval_combined_score": 0.7961937944750008,
+      "eval_f1": 0.8399366085578448,
+      "eval_loss": 0.5621684193611145,
+      "eval_runtime": 0.1459,
+      "eval_samples_per_second": 2795.865,
+      "eval_steps_per_second": 13.705,
+      "step": 225
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 4.69514799118042,
+      "learning_rate": 3.4000000000000007e-05,
+      "loss": 0.3506,
+      "step": 240
+    },
+    {
+      "epoch": 16.0,
+      "eval_accuracy": 0.7524509803921569,
+      "eval_combined_score": 0.7949050393104101,
+      "eval_f1": 0.8373590982286634,
+      "eval_loss": 0.5587612390518188,
+      "eval_runtime": 0.1412,
+      "eval_samples_per_second": 2889.737,
+      "eval_steps_per_second": 14.165,
+      "step": 240
+    },
+    {
+      "epoch": 16.0,
+      "step": 240,
+      "total_flos": 73027538583552.0,
+      "train_loss": 0.490890504916509,
+      "train_runtime": 29.3983,
+      "train_samples_per_second": 6238.45,
+      "train_steps_per_second": 25.512
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 750,
+  "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": 73027538583552.0,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}