<!-- 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. -->
ntu_adl_summarization_mt5_s
This model is a fine-tuned version of google/mt5-small on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 3.6583
- Rouge-1: 21.9729
- Rouge-2: 7.6735
- Rouge-l: 19.7497
- Ave Gen Len: 17.3098
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
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge-1 | Rouge-2 | Rouge-l | Ave Gen Len |
|---|---|---|---|---|---|---|---|
| 5.4447 | 1.0 | 1357 | 4.1235 | 17.7916 | 5.9785 | 16.5599 | 12.7161 |
| 4.7463 | 2.0 | 2714 | 3.9569 | 19.6608 | 6.7631 | 18.0768 | 14.8245 |
| 4.5203 | 3.0 | 4071 | 3.8545 | 20.5626 | 7.0737 | 18.7628 | 16.3307 |
| 4.4285 | 4.0 | 5428 | 3.7825 | 21.0690 | 7.2030 | 19.0863 | 16.7841 |
| 4.3196 | 5.0 | 6785 | 3.7269 | 21.2881 | 7.3307 | 19.2588 | 16.9276 |
| 4.2662 | 6.0 | 8142 | 3.7027 | 21.5793 | 7.5122 | 19.4806 | 17.0333 |
| 4.2057 | 7.0 | 9499 | 3.6764 | 21.7949 | 7.5987 | 19.6082 | 17.1811 |
| 4.1646 | 8.0 | 10856 | 3.6671 | 21.8164 | 7.5705 | 19.6207 | 17.2550 |
| 4.1399 | 9.0 | 12213 | 3.6602 | 21.9381 | 7.6577 | 19.7089 | 17.3014 |
| 4.1479 | 10.0 | 13570 | 3.6583 | 21.9729 | 7.6735 | 19.7497 | 17.3098 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1
