Edit model card

vit5-base-vietnews-summarization-finetuned-VN

This model is a fine-tuned version of VietAI/vit5-base-vietnews-summarization on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 1.8974
  • Rouge1: 46.8897
  • Rouge2: 21.2655
  • Rougel: 33.489
  • Rougelsum: 33.5578
  • Gen Len: 18.9871

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: 4
  • eval_batch_size: 4
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 16
  • 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 Rouge1 Rouge2 Rougel Rougelsum Gen Len
2.408 1.0 737 1.9329 46.6812 20.8232 33.1637 33.2377 18.9608
2.0444 2.0 1475 1.8841 46.7693 21.3121 33.37 33.4394 18.9623
1.7589 3.0 2212 1.8840 46.6303 21.0522 33.2678 33.3244 18.9777
1.6919 4.0 2950 1.8864 46.7928 21.395 33.3749 33.4439 18.9911
1.5844 5.0 3685 1.8974 46.8897 21.2655 33.489 33.5578 18.9871

Framework versions

  • Transformers 4.30.0
  • Pytorch 2.3.0+cu121
  • Datasets 2.19.2
  • Tokenizers 0.13.3
Downloads last month
103