mt5-small-finetuned-tradition-zh
This model is a fine-tuned version of google/mt5-small on the xlsum dataset. It achieves the following results on the evaluation set:
- Loss: 2.9218
- Rouge1: 5.7806
- Rouge2: 1.266
- Rougel: 5.761
- Rougelsum: 5.7833
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: 5.6e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
|---|---|---|---|---|---|---|---|
| 4.542 | 1.0 | 2336 | 3.1979 | 4.8334 | 1.025 | 4.8142 | 4.8326 |
| 3.7542 | 2.0 | 4672 | 3.0662 | 5.2155 | 1.0978 | 5.2025 | 5.2158 |
| 3.5706 | 3.0 | 7008 | 3.0070 | 5.5471 | 1.3397 | 5.5386 | 5.5391 |
| 3.4668 | 4.0 | 9344 | 2.9537 | 5.5865 | 1.1558 | 5.5816 | 5.5964 |
| 3.4082 | 5.0 | 11680 | 2.9391 | 5.8061 | 1.3462 | 5.7944 | 5.812 |
| 3.375 | 6.0 | 14016 | 2.9218 | 5.7806 | 1.266 | 5.761 | 5.7833 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.12.0+cu113
- Datasets 2.3.2
- Tokenizers 0.12.1