End of training

README.md

@@ -1,14 +1,32 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/tiny_bert_km_20_v2
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 - f1
 model-index:
 - name: tiny_bert_km_20_v2_mrpc
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE MRPC
+      type: glue
+      args: mrpc
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.7058823529411765
+    - name: F1
+      type: f1
+      value: 0.8187311178247734
 ---
 
 <!-- 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. -->
 
 # tiny_bert_km_20_v2_mrpc
 
-This model is a fine-tuned version of [Hartunka/tiny_bert_km_20_v2](https://huggingface.co/Hartunka/tiny_bert_km_20_v2) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/tiny_bert_km_20_v2](https://huggingface.co/Hartunka/tiny_bert_km_20_v2) on the GLUE MRPC dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7322
-- Accuracy: 0.6691
-- F1: 0.7723
-- Combined Score: 0.7207
+- Loss: 0.6003
+- Accuracy: 0.7059
+- F1: 0.8187
+- Combined Score: 0.7623
 
 ## Model description
 

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.7058823529411765,
+    "eval_combined_score": 0.7623067353829749,
+    "eval_f1": 0.8187311178247734,
+    "eval_loss": 0.6002748012542725,
+    "eval_runtime": 0.1598,
+    "eval_samples": 408,
+    "eval_samples_per_second": 2552.905,
+    "eval_steps_per_second": 12.514,
+    "total_flos": 673316591603712.0,
+    "train_loss": 0.5263727415175665,
+    "train_runtime": 19.8344,
+    "train_samples": 3668,
+    "train_samples_per_second": 9246.551,
+    "train_steps_per_second": 37.813
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 7.0,
+    "eval_accuracy": 0.7058823529411765,
+    "eval_combined_score": 0.7623067353829749,
+    "eval_f1": 0.8187311178247734,
+    "eval_loss": 0.6002748012542725,
+    "eval_runtime": 0.1598,
+    "eval_samples": 408,
+    "eval_samples_per_second": 2552.905,
+    "eval_steps_per_second": 12.514
+}

logs/events.out.tfevents.1745278323.s_005_m.2836625.39

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:409a01dafc891b218a8fa11dd6ea17f43e0994a0e30ba9d65752a59ab54d1c79
+size 459

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 7.0,
+    "total_flos": 673316591603712.0,
+    "train_loss": 0.5263727415175665,
+    "train_runtime": 19.8344,
+    "train_samples": 3668,
+    "train_samples_per_second": 9246.551,
+    "train_steps_per_second": 37.813
+}

trainer_state.json

@@ -0,0 +1,178 @@
+{
+  "best_global_step": 30,
+  "best_metric": 0.6002748012542725,
+  "best_model_checkpoint": "tiny_bert_km_20_v2_mrpc/checkpoint-30",
+  "epoch": 7.0,
+  "eval_steps": 500,
+  "global_step": 105,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.8159173727035522,
+      "learning_rate": 4.9e-05,
+      "loss": 0.6363,
+      "step": 15
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6936274509803921,
+      "eval_combined_score": 0.7511016274810077,
+      "eval_f1": 0.8085758039816233,
+      "eval_loss": 0.6099182963371277,
+      "eval_runtime": 0.1449,
+      "eval_samples_per_second": 2814.918,
+      "eval_steps_per_second": 13.799,
+      "step": 15
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.354114294052124,
+      "learning_rate": 4.8e-05,
+      "loss": 0.6012,
+      "step": 30
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.7058823529411765,
+      "eval_combined_score": 0.7623067353829749,
+      "eval_f1": 0.8187311178247734,
+      "eval_loss": 0.6002748012542725,
+      "eval_runtime": 0.1514,
+      "eval_samples_per_second": 2695.139,
+      "eval_steps_per_second": 13.211,
+      "step": 30
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.0689122676849365,
+      "learning_rate": 4.7e-05,
+      "loss": 0.5689,
+      "step": 45
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.7083333333333334,
+      "eval_combined_score": 0.7630487493619194,
+      "eval_f1": 0.8177641653905053,
+      "eval_loss": 0.60498046875,
+      "eval_runtime": 0.1712,
+      "eval_samples_per_second": 2382.829,
+      "eval_steps_per_second": 11.681,
+      "step": 45
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.1921436786651611,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.5499,
+      "step": 60
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7058823529411765,
+      "eval_combined_score": 0.7594832325453547,
+      "eval_f1": 0.8130841121495327,
+      "eval_loss": 0.6044909954071045,
+      "eval_runtime": 0.1595,
+      "eval_samples_per_second": 2557.85,
+      "eval_steps_per_second": 12.538,
+      "step": 60
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.2511504888534546,
+      "learning_rate": 4.5e-05,
+      "loss": 0.5072,
+      "step": 75
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6911764705882353,
+      "eval_combined_score": 0.7443020616606771,
+      "eval_f1": 0.797427652733119,
+      "eval_loss": 0.6230145692825317,
+      "eval_runtime": 0.1295,
+      "eval_samples_per_second": 3151.475,
+      "eval_steps_per_second": 15.448,
+      "step": 75
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.6471470594406128,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.4501,
+      "step": 90
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6715686274509803,
+      "eval_combined_score": 0.7222249916915919,
+      "eval_f1": 0.7728813559322034,
+      "eval_loss": 0.6497312188148499,
+      "eval_runtime": 0.1568,
+      "eval_samples_per_second": 2601.273,
+      "eval_steps_per_second": 12.751,
+      "step": 90
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 2.0113654136657715,
+      "learning_rate": 4.3e-05,
+      "loss": 0.371,
+      "step": 105
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6691176470588235,
+      "eval_combined_score": 0.7207308302747744,
+      "eval_f1": 0.7723440134907251,
+      "eval_loss": 0.7322385907173157,
+      "eval_runtime": 0.1344,
+      "eval_samples_per_second": 3036.525,
+      "eval_steps_per_second": 14.885,
+      "step": 105
+    },
+    {
+      "epoch": 7.0,
+      "step": 105,
+      "total_flos": 673316591603712.0,
+      "train_loss": 0.5263727415175665,
+      "train_runtime": 19.8344,
+      "train_samples_per_second": 9246.551,
+      "train_steps_per_second": 37.813
+    }
+  ],
+  "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": 673316591603712.0,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}