Edit model card

bart-large-cnn-finetuned-roundup-3-8

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

  • Loss: 1.4132
  • Rouge1: 49.6606
  • Rouge2: 28.4044
  • Rougel: 31.5419
  • Rougelsum: 46.2463
  • Gen Len: 142.0

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: 2
  • eval_batch_size: 2
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 8
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Rouge1 Rouge2 Rougel Rougelsum Gen Len
No log 1.0 258 1.2686 48.8513 28.7007 31.1199 45.7318 142.0
1.1738 2.0 516 1.1884 49.8072 28.9817 31.3611 46.9639 141.6875
1.1738 3.0 774 1.1970 49.3865 28.3426 30.0945 46.4681 141.3438
0.7069 4.0 1032 1.1984 50.6743 29.4728 31.5364 47.989 141.7188
0.7069 5.0 1290 1.2494 49.4461 28.9295 31.0334 46.6611 142.0
0.4618 6.0 1548 1.2954 50.6789 30.2783 32.1932 47.5929 142.0
0.4618 7.0 1806 1.3638 49.9476 30.223 32.4346 46.7383 142.0
0.3293 8.0 2064 1.4132 49.6606 28.4044 31.5419 46.2463 142.0

Framework versions

  • Transformers 4.18.0
  • Pytorch 1.11.0+cu113
  • Datasets 2.1.0
  • Tokenizers 0.12.1
Downloads last month
9