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_qqp
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE QQP
+      type: glue
+      args: qqp
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.8694286420974524
+    - name: F1
+      type: f1
+      value: 0.8237689868135536
 ---
 
 <!-- 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_qqp
 
-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 QQP dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3194
-- Accuracy: 0.8738
-- F1: 0.8306
-- Combined Score: 0.8522
+- Loss: 0.3084
+- Accuracy: 0.8694
+- F1: 0.8238
+- Combined Score: 0.8466
 
 ## Model description
 

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 15.0,
+    "eval_accuracy": 0.8694286420974524,
+    "eval_combined_score": 0.846598814455503,
+    "eval_f1": 0.8237689868135536,
+    "eval_loss": 0.3083794116973877,
+    "eval_runtime": 14.4623,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 2795.539,
+    "eval_steps_per_second": 10.925,
+    "total_flos": 6791195253749760.0,
+    "train_loss": 0.2819004479116305,
+    "train_runtime": 2527.15,
+    "train_samples": 363846,
+    "train_samples_per_second": 7198.742,
+    "train_steps_per_second": 28.134
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 15.0,
+    "eval_accuracy": 0.8694286420974524,
+    "eval_combined_score": 0.846598814455503,
+    "eval_f1": 0.8237689868135536,
+    "eval_loss": 0.3083794116973877,
+    "eval_runtime": 14.4623,
+    "eval_samples": 40430,
+    "eval_samples_per_second": 2795.539,
+    "eval_steps_per_second": 10.925
+}

logs/events.out.tfevents.1733338428.ki-g0008.1764072.25

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:344d2a086ab4069bfd742b2e81833611d610694ee335bc552227fffb93bf51da
+size 523

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 15.0,
+    "total_flos": 6791195253749760.0,
+    "train_loss": 0.2819004479116305,
+    "train_runtime": 2527.15,
+    "train_samples": 363846,
+    "train_samples_per_second": 7198.742,
+    "train_steps_per_second": 28.134
+}

trainer_state.json

@@ -0,0 +1,321 @@
+{
+  "best_metric": 0.3083794116973877,
+  "best_model_checkpoint": "bert_uncased_L-4_H-128_A-2_qqp/checkpoint-14220",
+  "epoch": 15.0,
+  "eval_steps": 500,
+  "global_step": 21330,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 2.0942301750183105,
+      "learning_rate": 4.9e-05,
+      "loss": 0.448,
+      "step": 1422
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.8203561711600297,
+      "eval_combined_score": 0.7969590676532806,
+      "eval_f1": 0.7735619641465316,
+      "eval_loss": 0.37850263714790344,
+      "eval_runtime": 13.8603,
+      "eval_samples_per_second": 2916.957,
+      "eval_steps_per_second": 11.399,
+      "step": 1422
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 3.4555797576904297,
+      "learning_rate": 4.8e-05,
+      "loss": 0.3757,
+      "step": 2844
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8351224338362602,
+      "eval_combined_score": 0.8165489730097775,
+      "eval_f1": 0.7979755121832949,
+      "eval_loss": 0.35622575879096985,
+      "eval_runtime": 14.0473,
+      "eval_samples_per_second": 2878.124,
+      "eval_steps_per_second": 11.248,
+      "step": 2844
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 2.538503646850586,
+      "learning_rate": 4.7e-05,
+      "loss": 0.3453,
+      "step": 4266
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.8520652980460054,
+      "eval_combined_score": 0.8286444313527647,
+      "eval_f1": 0.8052235646595239,
+      "eval_loss": 0.32627713680267334,
+      "eval_runtime": 13.8697,
+      "eval_samples_per_second": 2914.989,
+      "eval_steps_per_second": 11.392,
+      "step": 4266
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 4.184642314910889,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.3226,
+      "step": 5688
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8560474894880039,
+      "eval_combined_score": 0.8325137014198587,
+      "eval_f1": 0.8089799133517134,
+      "eval_loss": 0.3198220431804657,
+      "eval_runtime": 13.9346,
+      "eval_samples_per_second": 2901.408,
+      "eval_steps_per_second": 11.339,
+      "step": 5688
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 3.920213460922241,
+      "learning_rate": 4.5e-05,
+      "loss": 0.3049,
+      "step": 7110
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.8551817956962652,
+      "eval_combined_score": 0.8355282543687779,
+      "eval_f1": 0.8158747130412906,
+      "eval_loss": 0.3244872987270355,
+      "eval_runtime": 13.8957,
+      "eval_samples_per_second": 2909.543,
+      "eval_steps_per_second": 11.37,
+      "step": 7110
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 3.159956693649292,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.2903,
+      "step": 8532
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.8601533514716795,
+      "eval_combined_score": 0.8404501460294167,
+      "eval_f1": 0.8207469405871538,
+      "eval_loss": 0.3135942816734314,
+      "eval_runtime": 14.0266,
+      "eval_samples_per_second": 2882.384,
+      "eval_steps_per_second": 11.264,
+      "step": 8532
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 3.239968776702881,
+      "learning_rate": 4.3e-05,
+      "loss": 0.277,
+      "step": 9954
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.8602770220133564,
+      "eval_combined_score": 0.8401146066242081,
+      "eval_f1": 0.8199521912350598,
+      "eval_loss": 0.3187676668167114,
+      "eval_runtime": 13.7148,
+      "eval_samples_per_second": 2947.903,
+      "eval_steps_per_second": 11.52,
+      "step": 9954
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 2.6510915756225586,
+      "learning_rate": 4.2e-05,
+      "loss": 0.2667,
+      "step": 11376
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.864828097947069,
+      "eval_combined_score": 0.8452890928150039,
+      "eval_f1": 0.8257500876829386,
+      "eval_loss": 0.31460729241371155,
+      "eval_runtime": 14.029,
+      "eval_samples_per_second": 2881.878,
+      "eval_steps_per_second": 11.262,
+      "step": 11376
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 3.9660000801086426,
+      "learning_rate": 4.1e-05,
+      "loss": 0.2551,
+      "step": 12798
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.861192184021766,
+      "eval_combined_score": 0.8426061265044547,
+      "eval_f1": 0.8240200689871433,
+      "eval_loss": 0.3238842785358429,
+      "eval_runtime": 13.8769,
+      "eval_samples_per_second": 2913.471,
+      "eval_steps_per_second": 11.386,
+      "step": 12798
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 3.2269535064697266,
+      "learning_rate": 4e-05,
+      "loss": 0.2444,
+      "step": 14220
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.8694286420974524,
+      "eval_combined_score": 0.846598814455503,
+      "eval_f1": 0.8237689868135536,
+      "eval_loss": 0.3083794116973877,
+      "eval_runtime": 13.6982,
+      "eval_samples_per_second": 2951.479,
+      "eval_steps_per_second": 11.534,
+      "step": 14220
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 3.97849178314209,
+      "learning_rate": 3.9000000000000006e-05,
+      "loss": 0.2358,
+      "step": 15642
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.8668810289389067,
+      "eval_combined_score": 0.8482553292842682,
+      "eval_f1": 0.8296296296296296,
+      "eval_loss": 0.32236889004707336,
+      "eval_runtime": 13.9068,
+      "eval_samples_per_second": 2907.205,
+      "eval_steps_per_second": 11.361,
+      "step": 15642
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 2.4075608253479004,
+      "learning_rate": 3.8e-05,
+      "loss": 0.2276,
+      "step": 17064
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.8698738560474895,
+      "eval_combined_score": 0.8506017031191248,
+      "eval_f1": 0.83132955019076,
+      "eval_loss": 0.3146681487560272,
+      "eval_runtime": 13.9222,
+      "eval_samples_per_second": 2903.986,
+      "eval_steps_per_second": 11.349,
+      "step": 17064
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 4.484858989715576,
+      "learning_rate": 3.7e-05,
+      "loss": 0.2191,
+      "step": 18486
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.8707642839475637,
+      "eval_combined_score": 0.8519823015907012,
+      "eval_f1": 0.8332003192338387,
+      "eval_loss": 0.31845706701278687,
+      "eval_runtime": 13.8228,
+      "eval_samples_per_second": 2924.882,
+      "eval_steps_per_second": 11.43,
+      "step": 18486
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 4.709075927734375,
+      "learning_rate": 3.6e-05,
+      "loss": 0.2111,
+      "step": 19908
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.8720504575810042,
+      "eval_combined_score": 0.8528927761179858,
+      "eval_f1": 0.8337350946549673,
+      "eval_loss": 0.32821860909461975,
+      "eval_runtime": 14.1205,
+      "eval_samples_per_second": 2863.217,
+      "eval_steps_per_second": 11.189,
+      "step": 19908
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 4.495522499084473,
+      "learning_rate": 3.5e-05,
+      "loss": 0.2048,
+      "step": 21330
+    },
+    {
+      "epoch": 15.0,
+      "eval_accuracy": 0.8738313133811526,
+      "eval_combined_score": 0.8522011384722219,
+      "eval_f1": 0.8305709635632911,
+      "eval_loss": 0.31938907504081726,
+      "eval_runtime": 14.3581,
+      "eval_samples_per_second": 2815.826,
+      "eval_steps_per_second": 11.004,
+      "step": 21330
+    },
+    {
+      "epoch": 15.0,
+      "step": 21330,
+      "total_flos": 6791195253749760.0,
+      "train_loss": 0.2819004479116305,
+      "train_runtime": 2527.15,
+      "train_samples_per_second": 7198.742,
+      "train_steps_per_second": 28.134
+    }
+  ],
+  "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": 6791195253749760.0,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}