TH_mt5-base-finetuned-wikisql
This model is a fine-tuned version of google/mt5-base on the wikisql dataset. It achieves the following results on the evaluation set:
- Loss: 0.3517
- Rouge2 Precision: 0.6349
- Rouge2 Recall: 0.5656
- Rouge2 Fmeasure: 0.5929
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: 5e-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: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge2 Precision | Rouge2 Recall | Rouge2 Fmeasure |
|---|---|---|---|---|---|---|
| 0.4981 | 1.0 | 4049 | 0.4153 | 0.5813 | 0.5159 | 0.5409 |
| 0.4242 | 2.0 | 8098 | 0.3777 | 0.6117 | 0.543 | 0.5695 |
| 0.384 | 3.0 | 12147 | 0.3599 | 0.6264 | 0.5573 | 0.5844 |
| 0.3651 | 4.0 | 16196 | 0.3537 | 0.6343 | 0.5639 | 0.5917 |
| 0.3557 | 5.0 | 20245 | 0.3517 | 0.6349 | 0.5656 | 0.5929 |
Framework versions
- Transformers 4.26.1
- Pytorch 2.0.1+cu117
- Datasets 2.14.7.dev0
- Tokenizers 0.13.3