<!-- 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. -->
hyunwoongko-kobart-eb-finetuned-papers-meetings
This model is a fine-tuned version of hyunwoongko/kobart on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3136
- Rouge1: 18.3166
- Rouge2: 8.0509
- Rougel: 18.3332
- Rougelsum: 18.3146
- Gen Len: 19.9143
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: 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: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| 0.2118 | 1.0 | 7739 | 0.2951 | 18.0837 | 7.9585 | 18.0787 | 18.0784 | 19.896 |
| 0.1598 | 2.0 | 15478 | 0.2812 | 18.529 | 7.9891 | 18.5421 | 18.5271 | 19.8977 |
| 0.1289 | 3.0 | 23217 | 0.2807 | 18.0638 | 7.8086 | 18.0787 | 18.0583 | 19.9129 |
| 0.0873 | 4.0 | 30956 | 0.2923 | 18.3483 | 8.0233 | 18.3716 | 18.3696 | 19.914 |
| 0.0844 | 5.0 | 38695 | 0.3136 | 18.3166 | 8.0509 | 18.3332 | 18.3146 | 19.9143 |
Framework versions
- Transformers 4.19.4
- Pytorch 1.11.0+cu113
- Datasets 2.3.2
- Tokenizers 0.12.1
