AraBART-finetuned-wiki-ar

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

  • Loss: 2.4030
  • Rouge1: 0.9862
  • Rouge2: 0.2292
  • Rougel: 0.9902
  • Rougelsum: 0.9847
  • Gen Len: 19.3511

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: 8
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 10
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Rouge1 Rouge2 Rougel Rougelsum Gen Len
2.8633 1.0 2556 2.5599 0.7861 0.1289 0.7656 0.7721 19.2354
2.6525 2.0 5112 2.4824 0.7315 0.2374 0.7224 0.7357 19.261
2.5068 3.0 7668 2.4404 0.7772 0.2114 0.7671 0.7861 19.3035
2.4251 4.0 10224 2.4269 0.7464 0.2156 0.745 0.7504 19.2929
2.3739 5.0 12780 2.4119 0.7642 0.1879 0.7729 0.7774 19.3573
2.275 6.0 15336 2.4039 0.9048 0.1952 0.9198 0.9189 19.37
2.2787 7.0 17892 2.4007 0.9913 0.2278 0.9951 1.0038 19.335
2.2142 8.0 20448 2.4073 0.9736 0.238 0.9697 0.9773 19.3556
2.1852 9.0 23004 2.4007 0.9825 0.2322 0.9891 0.9962 19.3213
2.1597 10.0 25560 2.4030 0.9862 0.2292 0.9902 0.9847 19.3511

Framework versions

  • Transformers 4.25.1
  • Pytorch 1.13.0+cu116
  • Datasets 2.7.1
  • Tokenizers 0.13.2
Downloads last month
195
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for Jezia/AraBART-finetuned-wiki-ar

Finetunes
2 models