128Bert / README.md
IvashinMaxim's picture
End of training
9914629 verified
metadata
license: mit
base_model: cointegrated/rubert-tiny2
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: 128Bert
    results: []

128Bert

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: 0.8346
  • Accuracy: 0.7033

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: 1e-05
  • train_batch_size: 128
  • eval_batch_size: 128
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Accuracy
1.1934 1.0 2074 1.1488 0.6027
1.0626 2.0 4148 1.0247 0.6459
0.9729 3.0 6222 0.9483 0.6658
0.908 4.0 8296 0.9041 0.6811
0.8684 5.0 10370 0.8771 0.6897
0.8348 6.0 12444 0.8593 0.6956
0.8055 7.0 14518 0.8507 0.6991
0.7924 8.0 16592 0.8410 0.7017
0.7857 9.0 18666 0.8349 0.7037
0.7732 10.0 20740 0.8346 0.7033

Framework versions

  • Transformers 4.40.1
  • Pytorch 2.2.1+cu121
  • Datasets 2.19.0
  • Tokenizers 0.19.1