Edit model card

pegasus-xsum-finetuned-cnn_dailymail

This model is a fine-tuned version of google/pegasus-xsum on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.8958
  • Rouge1: 45.7795
  • Rouge2: 23.3182
  • Rougel: 32.9241
  • Rougelsum: 42.3126
  • Bleu 1: 35.4715
  • Bleu 2: 24.0726
  • Bleu 3: 17.9591
  • Meteor: 32.8897
  • Lungime rezumat: 43.3773
  • Lungime original: 48.6937

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: 5.6e-05
  • train_batch_size: 4
  • eval_batch_size: 4
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 4

Training results

Training Loss Epoch Step Validation Loss Rouge1 Rouge2 Rougel Rougelsum Bleu 1 Bleu 2 Bleu 3 Meteor Lungime rezumat Lungime original
1.1281 1.0 14330 0.9373 44.64 22.2111 32.0228 41.1223 34.4946 23.079 17.0673 31.8685 43.543 48.6937
0.9091 2.0 28660 0.9095 45.0713 22.7428 32.4247 41.554 34.9397 23.5631 17.5094 32.1814 43.3467 48.6937
0.8455 3.0 42990 0.8982 45.5457 23.1315 32.7153 42.0349 35.2659 23.8773 17.8174 32.7185 43.5743 48.6937
0.8076 4.0 57320 0.8958 45.7795 23.3182 32.9241 42.3126 35.4715 24.0726 17.9591 32.8897 43.3773 48.6937

Framework versions

  • Transformers 4.40.0
  • Pytorch 2.2.2+cu118
  • Datasets 2.19.0
  • Tokenizers 0.19.1
Downloads last month
8
Safetensors
Model size
570M params
Tensor type
F32
·

Finetuned from