End of training

README.md

@@ -1,14 +1,32 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/tiny_bert_km_100_v2
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 - f1
 model-index:
 - name: tiny_bert_km_100_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: 0.6985294117647058
+    - name: F1
+      type: f1
+      value: 0.8056872037914692
 ---
 
 <!-- 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_100_v2_mrpc
 
-This model is a fine-tuned version of [Hartunka/tiny_bert_km_100_v2](https://huggingface.co/Hartunka/tiny_bert_km_100_v2) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/tiny_bert_km_100_v2](https://huggingface.co/Hartunka/tiny_bert_km_100_v2) on the GLUE MRPC dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.9051
-- Accuracy: 0.6495
-- F1: 0.7478
-- Combined Score: 0.6987
+- Loss: 0.5924
+- Accuracy: 0.6985
+- F1: 0.8057
+- Combined Score: 0.7521
 
 ## Model description
 

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 9.0,
+    "eval_accuracy": 0.6985294117647058,
+    "eval_combined_score": 0.7521083077780875,
+    "eval_f1": 0.8056872037914692,
+    "eval_loss": 0.5924485325813293,
+    "eval_runtime": 0.154,
+    "eval_samples": 408,
+    "eval_samples_per_second": 2648.878,
+    "eval_steps_per_second": 12.985,
+    "total_flos": 865692760633344.0,
+    "train_loss": 0.4507062558774595,
+    "train_runtime": 24.7934,
+    "train_samples": 3668,
+    "train_samples_per_second": 7397.122,
+    "train_steps_per_second": 30.25
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 9.0,
+    "eval_accuracy": 0.6985294117647058,
+    "eval_combined_score": 0.7521083077780875,
+    "eval_f1": 0.8056872037914692,
+    "eval_loss": 0.5924485325813293,
+    "eval_runtime": 0.154,
+    "eval_samples": 408,
+    "eval_samples_per_second": 2648.878,
+    "eval_steps_per_second": 12.985
+}

logs/events.out.tfevents.1745288231.s_005_m.2836625.75

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cffd9723b4396915b96f885fd8641668224e8573a11905ac4ca4effd88ee92c7
+size 467

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 9.0,
+    "total_flos": 865692760633344.0,
+    "train_loss": 0.4507062558774595,
+    "train_runtime": 24.7934,
+    "train_samples": 3668,
+    "train_samples_per_second": 7397.122,
+    "train_steps_per_second": 30.25
+}

trainer_state.json

@@ -0,0 +1,214 @@
+{
+  "best_global_step": 60,
+  "best_metric": 0.5924485325813293,
+  "best_model_checkpoint": "tiny_bert_km_100_v2_mrpc/checkpoint-60",
+  "epoch": 9.0,
+  "eval_steps": 500,
+  "global_step": 135,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.7998549938201904,
+      "learning_rate": 4.9e-05,
+      "loss": 0.6358,
+      "step": 15
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7058823529411765,
+      "eval_combined_score": 0.7631208171891511,
+      "eval_f1": 0.8203592814371258,
+      "eval_loss": 0.6057416200637817,
+      "eval_runtime": 0.145,
+      "eval_samples_per_second": 2814.34,
+      "eval_steps_per_second": 13.796,
+      "step": 15
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.222335696220398,
+      "learning_rate": 4.8e-05,
+      "loss": 0.5975,
+      "step": 30
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.7083333333333334,
+      "eval_combined_score": 0.7657565626547795,
+      "eval_f1": 0.8231797919762258,
+      "eval_loss": 0.5953512787818909,
+      "eval_runtime": 0.1347,
+      "eval_samples_per_second": 3029.043,
+      "eval_steps_per_second": 14.848,
+      "step": 30
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.0654611587524414,
+      "learning_rate": 4.7e-05,
+      "loss": 0.5692,
+      "step": 45
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.7083333333333334,
+      "eval_combined_score": 0.7649612693653174,
+      "eval_f1": 0.8215892053973014,
+      "eval_loss": 0.5976826548576355,
+      "eval_runtime": 0.1453,
+      "eval_samples_per_second": 2808.202,
+      "eval_steps_per_second": 13.766,
+      "step": 45
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.4689693450927734,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.5442,
+      "step": 60
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6985294117647058,
+      "eval_combined_score": 0.7521083077780875,
+      "eval_f1": 0.8056872037914692,
+      "eval_loss": 0.5924485325813293,
+      "eval_runtime": 0.1383,
+      "eval_samples_per_second": 2949.896,
+      "eval_steps_per_second": 14.46,
+      "step": 60
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.3699629306793213,
+      "learning_rate": 4.5e-05,
+      "loss": 0.4983,
+      "step": 75
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6985294117647058,
+      "eval_combined_score": 0.7499109094364724,
+      "eval_f1": 0.8012924071082391,
+      "eval_loss": 0.6211437582969666,
+      "eval_runtime": 0.1558,
+      "eval_samples_per_second": 2618.223,
+      "eval_steps_per_second": 12.834,
+      "step": 75
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.584648847579956,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.4324,
+      "step": 90
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6862745098039216,
+      "eval_combined_score": 0.7342937174869948,
+      "eval_f1": 0.782312925170068,
+      "eval_loss": 0.6470237970352173,
+      "eval_runtime": 0.1673,
+      "eval_samples_per_second": 2438.639,
+      "eval_steps_per_second": 11.954,
+      "step": 90
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 2.687011241912842,
+      "learning_rate": 4.3e-05,
+      "loss": 0.3495,
+      "step": 105
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.678921568627451,
+      "eval_combined_score": 0.7286316810988355,
+      "eval_f1": 0.7783417935702199,
+      "eval_loss": 0.7373393177986145,
+      "eval_runtime": 0.1321,
+      "eval_samples_per_second": 3089.682,
+      "eval_steps_per_second": 15.146,
+      "step": 105
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 1.913145661354065,
+      "learning_rate": 4.2e-05,
+      "loss": 0.2481,
+      "step": 120
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.678921568627451,
+      "eval_combined_score": 0.7308372353916691,
+      "eval_f1": 0.7827529021558872,
+      "eval_loss": 0.8263801336288452,
+      "eval_runtime": 0.1498,
+      "eval_samples_per_second": 2723.288,
+      "eval_steps_per_second": 13.349,
+      "step": 120
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 2.4080166816711426,
+      "learning_rate": 4.1e-05,
+      "loss": 0.1813,
+      "step": 135
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.6495098039215687,
+      "eval_combined_score": 0.6986526091918248,
+      "eval_f1": 0.7477954144620811,
+      "eval_loss": 0.9050607681274414,
+      "eval_runtime": 0.1354,
+      "eval_samples_per_second": 3012.778,
+      "eval_steps_per_second": 14.769,
+      "step": 135
+    },
+    {
+      "epoch": 9.0,
+      "step": 135,
+      "total_flos": 865692760633344.0,
+      "train_loss": 0.4507062558774595,
+      "train_runtime": 24.7934,
+      "train_samples_per_second": 7397.122,
+      "train_steps_per_second": 30.25
+    }
+  ],
+  "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": 865692760633344.0,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}