UNIST-Eunchan's picture
Update README.md
526a8d8
metadata
tags:
  - generated_from_trainer
model-index:
  - name: pegasus-x-booksum-chapter
    results: []

pegasus-x-booksum-chapter

This model is a fine-tuned version of google/pegasus-x-large on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 2.5993

{'rouge1': 0.23378525407555606,
'rouge2': 0.0362962105694859,
'rougeL': 0.13587636708639556,
'rougeLsum': 0.13593997471043634}

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: 5e-05
  • train_batch_size: 1
  • eval_batch_size: 1
  • seed: 42
  • gradient_accumulation_steps: 32
  • total_train_batch_size: 32
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 200
  • num_epochs: 8

Training results

Training Loss Epoch Step Validation Loss
2.8377 0.67 200 2.7300
2.7462 1.33 400 2.6613
2.7148 2.0 600 2.6345
2.6242 2.67 800 2.6249
2.5971 3.33 1000 2.6150
2.6103 4.0 1200 2.6092
2.5763 4.67 1400 2.6083
2.5737 5.33 1600 2.6035
2.6252 6.0 1800 2.6007
2.5402 6.67 2000 2.6004
2.5278 7.33 2200 2.6007
2.5173 8.0 2400 2.5993

Framework versions

  • Transformers 4.30.2
  • Pytorch 2.0.1+cu117
  • Datasets 2.13.1
  • Tokenizers 0.13.3