tringuyen-uit's picture
End of training
113ad32 verified
metadata
license: mit
base_model: VietAI/vit5-base
tags:
  - generated_from_trainer
metrics:
  - f1
model-index:
  - name: MRC_ER_vit5-base_syl_ViWikiFC
    results: []

MRC_ER_vit5-base_syl_ViWikiFC

This model is a fine-tuned version of VietAI/vit5-base on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 3.6043
  • Exact Match: 0.8057
  • F1: 0.8304

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
  • 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 Exact Match F1
0.5884 1.0 2093 1.6800 0.7866 0.8201
0.4403 2.0 4186 1.7782 0.8029 0.8295
0.2864 3.0 6279 2.1889 0.8067 0.8301
0.2193 4.0 8372 3.0962 0.8096 0.8337
0.1426 5.0 10465 3.6043 0.8057 0.8304

Framework versions

  • Transformers 4.39.3
  • Pytorch 2.1.2
  • Datasets 2.18.0
  • Tokenizers 0.15.2