Ilya-Nazimov's picture
End of training
0f4f25e verified
metadata
license: mit
base_model: cointegrated/rubert-tiny2
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: rubert-tiny2-odonata-extended-305-1-ner
    results: []

rubert-tiny2-odonata-extended-305-1-ner

This model is a fine-tuned version of cointegrated/rubert-tiny2 on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0171
  • Precision: 0.5782
  • Recall: 0.6289
  • F1: 0.6025
  • Accuracy: 0.9940

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 2e-05
  • train_batch_size: 16
  • eval_batch_size: 16
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 30

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
No log 1.0 25 0.2514 0.0 0.0 0.0 0.9913
No log 2.0 50 0.0660 0.0 0.0 0.0 0.9913
No log 3.0 75 0.0579 0.0 0.0 0.0 0.9913
No log 4.0 100 0.0557 0.0 0.0 0.0 0.9913
No log 5.0 125 0.0528 0.0 0.0 0.0 0.9913
No log 6.0 150 0.0465 0.0 0.0 0.0 0.9913
No log 7.0 175 0.0359 1.0 0.0052 0.0103 0.9914
No log 8.0 200 0.0278 0.5802 0.3918 0.4677 0.9921
No log 9.0 225 0.0241 0.5940 0.4072 0.4832 0.9922
No log 10.0 250 0.0223 0.6 0.4175 0.4924 0.9925
No log 11.0 275 0.0212 0.5417 0.4691 0.5028 0.9930
No log 12.0 300 0.0204 0.52 0.4691 0.4932 0.9931
No log 13.0 325 0.0199 0.5579 0.5464 0.5521 0.9936
No log 14.0 350 0.0194 0.5761 0.5464 0.5608 0.9939
No log 15.0 375 0.0190 0.5761 0.5464 0.5608 0.9938
No log 16.0 400 0.0187 0.5670 0.5670 0.5670 0.9939
No log 17.0 425 0.0184 0.5685 0.5773 0.5729 0.9938
No log 18.0 450 0.0182 0.5707 0.6031 0.5865 0.9939
No log 19.0 475 0.0180 0.5680 0.6031 0.5850 0.9940
0.0748 20.0 500 0.0177 0.5764 0.6031 0.5894 0.9941
0.0748 21.0 525 0.0176 0.5907 0.5876 0.5891 0.9941
0.0748 22.0 550 0.0176 0.5769 0.6186 0.5970 0.9941
0.0748 23.0 575 0.0174 0.5939 0.6031 0.5985 0.9942
0.0748 24.0 600 0.0173 0.5854 0.6186 0.6015 0.9942
0.0748 25.0 625 0.0172 0.5902 0.6237 0.6065 0.9942
0.0748 26.0 650 0.0172 0.5865 0.6289 0.6070 0.9942
0.0748 27.0 675 0.0171 0.5862 0.6134 0.5995 0.9941
0.0748 28.0 700 0.0171 0.5882 0.6186 0.6030 0.9942
0.0748 29.0 725 0.0171 0.5817 0.6237 0.6020 0.9941
0.0748 30.0 750 0.0171 0.5782 0.6289 0.6025 0.9940

Framework versions

  • Transformers 4.41.2
  • Pytorch 2.3.1+cpu
  • Datasets 2.19.2
  • Tokenizers 0.19.1