End of training

README.md

@@ -1,14 +1,32 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/tiny_bert_km_10_v1
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 - f1
 model-index:
 - name: tiny_bert_km_10_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: 0.7009803921568627
+    - name: F1
+      type: f1
+      value: 0.8032258064516129
 ---
 
 <!-- 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_10_v1_mrpc
 
-This model is a fine-tuned version of [Hartunka/tiny_bert_km_10_v1](https://huggingface.co/Hartunka/tiny_bert_km_10_v1) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/tiny_bert_km_10_v1](https://huggingface.co/Hartunka/tiny_bert_km_10_v1) on the GLUE MRPC dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0956
-- Accuracy: 0.6225
-- F1: 0.7148
-- Combined Score: 0.6687
+- Loss: 0.5893
+- Accuracy: 0.7010
+- F1: 0.8032
+- Combined Score: 0.7521
 
 ## Model description
 

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 9.0,
+    "eval_accuracy": 0.7009803921568627,
+    "eval_combined_score": 0.7521030993042378,
+    "eval_f1": 0.8032258064516129,
+    "eval_loss": 0.5893135666847229,
+    "eval_runtime": 0.1584,
+    "eval_samples": 408,
+    "eval_samples_per_second": 2575.322,
+    "eval_steps_per_second": 12.624,
+    "total_flos": 865692760633344.0,
+    "train_loss": 0.44955003173262986,
+    "train_runtime": 25.8086,
+    "train_samples": 3668,
+    "train_samples_per_second": 7106.171,
+    "train_steps_per_second": 29.06
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 9.0,
+    "eval_accuracy": 0.7009803921568627,
+    "eval_combined_score": 0.7521030993042378,
+    "eval_f1": 0.8032258064516129,
+    "eval_loss": 0.5893135666847229,
+    "eval_runtime": 0.1584,
+    "eval_samples": 408,
+    "eval_samples_per_second": 2575.322,
+    "eval_steps_per_second": 12.624
+}

logs/events.out.tfevents.1744633015.s_005_m.2701049.21

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c269051255b93fb333c64fa4710913b9855618a051b53885a1267bc7be8a4fb4
+size 467

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 9.0,
+    "total_flos": 865692760633344.0,
+    "train_loss": 0.44955003173262986,
+    "train_runtime": 25.8086,
+    "train_samples": 3668,
+    "train_samples_per_second": 7106.171,
+    "train_steps_per_second": 29.06
+}

trainer_state.json

@@ -0,0 +1,214 @@
+{
+  "best_global_step": 60,
+  "best_metric": 0.5893135666847229,
+  "best_model_checkpoint": "tiny_bert_km_10_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": 0.7777377963066101,
+      "learning_rate": 4.9e-05,
+      "loss": 0.6324,
+      "step": 15
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7107843137254902,
+      "eval_combined_score": 0.7665367351759982,
+      "eval_f1": 0.822289156626506,
+      "eval_loss": 0.6063352227210999,
+      "eval_runtime": 0.1334,
+      "eval_samples_per_second": 3057.75,
+      "eval_steps_per_second": 14.989,
+      "step": 15
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.288647174835205,
+      "learning_rate": 4.8e-05,
+      "loss": 0.596,
+      "step": 30
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.7034313725490197,
+      "eval_combined_score": 0.7612822034643454,
+      "eval_f1": 0.8191330343796711,
+      "eval_loss": 0.5933586359024048,
+      "eval_runtime": 0.1309,
+      "eval_samples_per_second": 3116.868,
+      "eval_steps_per_second": 15.279,
+      "step": 30
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.0193438529968262,
+      "learning_rate": 4.7e-05,
+      "loss": 0.5663,
+      "step": 45
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.7083333333333334,
+      "eval_combined_score": 0.7641515380736259,
+      "eval_f1": 0.8199697428139183,
+      "eval_loss": 0.5923270583152771,
+      "eval_runtime": 0.1384,
+      "eval_samples_per_second": 2947.473,
+      "eval_steps_per_second": 14.448,
+      "step": 45
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.2852709293365479,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.5449,
+      "step": 60
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7009803921568627,
+      "eval_combined_score": 0.7521030993042378,
+      "eval_f1": 0.8032258064516129,
+      "eval_loss": 0.5893135666847229,
+      "eval_runtime": 0.1389,
+      "eval_samples_per_second": 2937.399,
+      "eval_steps_per_second": 14.399,
+      "step": 60
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.8828601837158203,
+      "learning_rate": 4.5e-05,
+      "loss": 0.4957,
+      "step": 75
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6568627450980392,
+      "eval_combined_score": 0.7034313725490196,
+      "eval_f1": 0.75,
+      "eval_loss": 0.630436897277832,
+      "eval_runtime": 0.1304,
+      "eval_samples_per_second": 3128.67,
+      "eval_steps_per_second": 15.337,
+      "step": 75
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.855084776878357,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.4356,
+      "step": 90
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6936274509803921,
+      "eval_combined_score": 0.7403401309416441,
+      "eval_f1": 0.787052810902896,
+      "eval_loss": 0.6516110301017761,
+      "eval_runtime": 0.1294,
+      "eval_samples_per_second": 3153.752,
+      "eval_steps_per_second": 15.46,
+      "step": 90
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 2.397104501724243,
+      "learning_rate": 4.3e-05,
+      "loss": 0.3509,
+      "step": 105
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6887254901960784,
+      "eval_combined_score": 0.7397498950156669,
+      "eval_f1": 0.7907742998352554,
+      "eval_loss": 0.7439402341842651,
+      "eval_runtime": 0.1343,
+      "eval_samples_per_second": 3037.258,
+      "eval_steps_per_second": 14.889,
+      "step": 105
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 2.2138803005218506,
+      "learning_rate": 4.2e-05,
+      "loss": 0.2548,
+      "step": 120
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6617647058823529,
+      "eval_combined_score": 0.7085419274092616,
+      "eval_f1": 0.7553191489361702,
+      "eval_loss": 0.8599588871002197,
+      "eval_runtime": 0.1419,
+      "eval_samples_per_second": 2876.022,
+      "eval_steps_per_second": 14.098,
+      "step": 120
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 2.5679445266723633,
+      "learning_rate": 4.1e-05,
+      "loss": 0.1693,
+      "step": 135
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.6225490196078431,
+      "eval_combined_score": 0.668681917211329,
+      "eval_f1": 0.7148148148148148,
+      "eval_loss": 1.095586895942688,
+      "eval_runtime": 0.1285,
+      "eval_samples_per_second": 3174.615,
+      "eval_steps_per_second": 15.562,
+      "step": 135
+    },
+    {
+      "epoch": 9.0,
+      "step": 135,
+      "total_flos": 865692760633344.0,
+      "train_loss": 0.44955003173262986,
+      "train_runtime": 25.8086,
+      "train_samples_per_second": 7106.171,
+      "train_steps_per_second": 29.06
+    }
+  ],
+  "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
+}