cuongpp's picture
End of training
0583303 verified
metadata
base_model: vinai/bartpho-word-base
tags:
  - generated_from_trainer
model-index:
  - name: bartpho-word-base-ed-multi
    results: []

bartpho-word-base-ed-multi

This model is a fine-tuned version of vinai/bartpho-word-base on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0119
  • F1 Micro: 0.7410
  • Recall Micro: 0.6625
  • Precision Micro: 0.8405
  • F1 Macro: 0.1860
  • Recall Macro: 0.1842
  • Precision Macro: 0.1913

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: 8
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 32
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 5

Training results

Training Loss Epoch Step Validation Loss F1 Micro Recall Micro Precision Micro F1 Macro Recall Macro Precision Macro
No log 1.0 373 0.0157 0.7226 0.6524 0.8096 0.1610 0.1642 0.1658
0.1287 2.0 746 0.0127 0.7332 0.6505 0.8399 0.1715 0.1628 0.1853
0.0132 3.0 1119 0.0130 0.7253 0.6411 0.8351 0.1860 0.2003 0.1825
0.0132 4.0 1492 0.0118 0.7422 0.6782 0.8195 0.1746 0.1612 0.1978
0.0104 5.0 1865 0.0119 0.7410 0.6625 0.8405 0.1860 0.1842 0.1913

Framework versions

  • Transformers 4.41.2
  • Pytorch 2.3.0+cu121
  • Datasets 2.20.0
  • Tokenizers 0.19.1