End of training

README.md

@@ -1,13 +1,31 @@
 ---
+language:
+- en
 base_model: Hartunka/tiny_bert_km_50_v1
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 - f1
 model-index:
 - name: tiny_bert_km_50_v1_mrpc
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE MRPC
+      type: glue
+      args: mrpc
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.7034313725490197
+    - name: F1
+      type: f1
+      value: 0.8141321044546851
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -15,12 +33,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # tiny_bert_km_50_v1_mrpc
 
-This model is a fine-tuned version of [Hartunka/tiny_bert_km_50_v1](https://huggingface.co/Hartunka/tiny_bert_km_50_v1) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/tiny_bert_km_50_v1](https://huggingface.co/Hartunka/tiny_bert_km_50_v1) on the GLUE MRPC dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.9983
-- Accuracy: 0.6765
-- F1: 0.7617
-- Combined Score: 0.7191
+- Loss: 0.5948
+- Accuracy: 0.7034
+- F1: 0.8141
+- Combined Score: 0.7588
 
 ## Model description
 

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 9.0,
+    "eval_accuracy": 0.7034313725490197,
+    "eval_combined_score": 0.7587817385018524,
+    "eval_f1": 0.8141321044546851,
+    "eval_loss": 0.5948036909103394,
+    "eval_runtime": 0.217,
+    "eval_samples": 408,
+    "eval_samples_per_second": 1880.321,
+    "eval_steps_per_second": 9.217,
+    "total_flos": 865692760633344.0,
+    "train_loss": 0.4542141667118779,
+    "train_runtime": 31.7437,
+    "train_samples": 3668,
+    "train_samples_per_second": 5777.517,
+    "train_steps_per_second": 23.627
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 9.0,
+    "eval_accuracy": 0.7034313725490197,
+    "eval_combined_score": 0.7587817385018524,
+    "eval_f1": 0.8141321044546851,
+    "eval_loss": 0.5948036909103394,
+    "eval_runtime": 0.217,
+    "eval_samples": 408,
+    "eval_samples_per_second": 1880.321,
+    "eval_steps_per_second": 9.217
+}

logs/events.out.tfevents.1744812263.s_004_m

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b229660e12ec3e4e127b675b70a3c99237383972b5dc687c651b5dbc1998115
+size 40

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 9.0,
+    "total_flos": 865692760633344.0,
+    "train_loss": 0.4542141667118779,
+    "train_runtime": 31.7437,
+    "train_samples": 3668,
+    "train_samples_per_second": 5777.517,
+    "train_steps_per_second": 23.627
+}

trainer_state.json

@@ -0,0 +1,192 @@
+{
+  "best_metric": 0.5948036909103394,
+  "best_model_checkpoint": "tiny_bert_km_50_v1_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": 0.7772509455680847,
+      "learning_rate": 4.9e-05,
+      "loss": 0.6321,
+      "step": 15
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.696078431372549,
+      "eval_combined_score": 0.7523602033405955,
+      "eval_f1": 0.808641975308642,
+      "eval_loss": 0.6045879125595093,
+      "eval_runtime": 0.2802,
+      "eval_samples_per_second": 1456.004,
+      "eval_steps_per_second": 7.137,
+      "step": 15
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.1948821544647217,
+      "learning_rate": 4.8e-05,
+      "loss": 0.5989,
+      "step": 30
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6936274509803921,
+      "eval_combined_score": 0.7539459691751886,
+      "eval_f1": 0.8142644873699851,
+      "eval_loss": 0.6042695641517639,
+      "eval_runtime": 0.1803,
+      "eval_samples_per_second": 2263.398,
+      "eval_steps_per_second": 11.095,
+      "step": 30
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.9500348567962646,
+      "learning_rate": 4.7e-05,
+      "loss": 0.5748,
+      "step": 45
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.7009803921568627,
+      "eval_combined_score": 0.7597163865546218,
+      "eval_f1": 0.8184523809523809,
+      "eval_loss": 0.5988913774490356,
+      "eval_runtime": 0.1664,
+      "eval_samples_per_second": 2452.036,
+      "eval_steps_per_second": 12.02,
+      "step": 45
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.1909563541412354,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.5524,
+      "step": 60
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7034313725490197,
+      "eval_combined_score": 0.7587817385018524,
+      "eval_f1": 0.8141321044546851,
+      "eval_loss": 0.5948036909103394,
+      "eval_runtime": 0.1699,
+      "eval_samples_per_second": 2400.82,
+      "eval_steps_per_second": 11.769,
+      "step": 60
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.895203948020935,
+      "learning_rate": 4.5e-05,
+      "loss": 0.5052,
+      "step": 75
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6936274509803921,
+      "eval_combined_score": 0.7435079403662291,
+      "eval_f1": 0.7933884297520661,
+      "eval_loss": 0.6062532067298889,
+      "eval_runtime": 0.1685,
+      "eval_samples_per_second": 2420.962,
+      "eval_steps_per_second": 11.867,
+      "step": 75
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.7124041318893433,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.4327,
+      "step": 90
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6887254901960784,
+      "eval_combined_score": 0.7331543387933106,
+      "eval_f1": 0.7775831873905429,
+      "eval_loss": 0.6554005742073059,
+      "eval_runtime": 0.1673,
+      "eval_samples_per_second": 2438.775,
+      "eval_steps_per_second": 11.955,
+      "step": 90
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 4.20126485824585,
+      "learning_rate": 4.3e-05,
+      "loss": 0.3584,
+      "step": 105
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.7058823529411765,
+      "eval_combined_score": 0.7491349480968859,
+      "eval_f1": 0.7923875432525952,
+      "eval_loss": 0.7306780815124512,
+      "eval_runtime": 0.1743,
+      "eval_samples_per_second": 2341.225,
+      "eval_steps_per_second": 11.477,
+      "step": 105
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 2.7051854133605957,
+      "learning_rate": 4.2e-05,
+      "loss": 0.258,
+      "step": 120
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6936274509803921,
+      "eval_combined_score": 0.7396096088521171,
+      "eval_f1": 0.7855917667238422,
+      "eval_loss": 0.8255677819252014,
+      "eval_runtime": 0.213,
+      "eval_samples_per_second": 1915.555,
+      "eval_steps_per_second": 9.39,
+      "step": 120
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 3.029175043106079,
+      "learning_rate": 4.1e-05,
+      "loss": 0.1754,
+      "step": 135
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.6764705882352942,
+      "eval_combined_score": 0.7191017201104268,
+      "eval_f1": 0.7617328519855595,
+      "eval_loss": 0.9982824921607971,
+      "eval_runtime": 0.1727,
+      "eval_samples_per_second": 2362.799,
+      "eval_steps_per_second": 11.582,
+      "step": 135
+    },
+    {
+      "epoch": 9.0,
+      "step": 135,
+      "total_flos": 865692760633344.0,
+      "train_loss": 0.4542141667118779,
+      "train_runtime": 31.7437,
+      "train_samples_per_second": 5777.517,
+      "train_steps_per_second": 23.627
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 750,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 50,
+  "save_steps": 500,
+  "total_flos": 865692760633344.0,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}