Edit model card

pegasus_cnn_news_article_title_12000_with_3_epochs

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

  • Loss: 0.1645

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: 16
  • total_train_batch_size: 16
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 500
  • num_epochs: 3

Training results

Training Loss Epoch Step Validation Loss
0.2874 0.65 500 0.2258
0.2286 1.31 1000 0.1843
0.211 1.96 1500 0.1699
0.18 2.61 2000 0.1645

Framework versions

  • Transformers 4.18.0
  • Pytorch 1.11.0
  • Datasets 2.1.0
  • Tokenizers 0.12.1
Downloads last month
1
Hosted inference API
This model can be loaded on the Inference API on-demand.