CoolFace
Modelpublic

eunbeee/ainize-kobart-news-eb-finetuned-meetings-papers

sourceHugging Facemitupdated 4y agoView on Hugging Face
1likes9downloads
Model Card

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

ainize-kobart-news-eb-finetuned-meetings-papers

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

  • —Loss: 0.3289
  • —Rouge1: 17.3988
  • —Rouge2: 7.0454
  • —Rougel: 17.3877
  • —Rougelsum: 17.42
  • —Gen Len: 19.9473

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
  • —trainbatchsize: 4
  • —evalbatchsize: 4
  • —seed: 42
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: linear
  • —num_epochs: 5
  • —mixedprecisiontraining: Native AMP

Training results

Training LossEpochStepValidation LossRouge1Rouge2RougelRougelsumGen Len
0.14021.075880.293017.14217.014117.121117.147319.9374
0.09972.0151760.284217.16926.882417.155717.198519.9435
0.06923.0227640.305217.42417.108317.402817.447219.9453
0.05564.0303520.328917.39887.045417.387717.4219.9473
0.05335.0379400.328917.39887.045417.387717.4219.9473

Framework versions

  • —Transformers 4.19.4
  • —Pytorch 1.11.0+cu113
  • —Datasets 2.2.2
  • —Tokenizers 0.12.1