<!-- 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. -->
mt5-base-coba
This model is a fine-tuned version of google/mt5-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5870
- Rouge1: 0.4338
- Rouge2: 0.2876
- Rougel: 0.3743
- Rougelsum: 0.409
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-06
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
|---|---|---|---|---|---|---|---|
| 7.0922 | 1.0 | 7452 | 0.6538 | 0.3566 | 0.239 | 0.3218 | 0.3348 |
| 0.9442 | 2.0 | 14904 | 0.6900 | 0.427 | 0.2868 | 0.3711 | 0.402 |
| 3.0789 | 3.0 | 22356 | 0.6775 | 0.3808 | 0.2584 | 0.3398 | 0.3567 |
| 1.0565 | 4.0 | 29808 | 0.5928 | 0.4348 | 0.2882 | 0.3756 | 0.4096 |
| 0.7872 | 5.0 | 37260 | 0.5870 | 0.4338 | 0.2876 | 0.3743 | 0.409 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.9.1
- Datasets 2.7.1
- Tokenizers 0.13.2
