<!-- 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-small-finetuned-14feb-1
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: 2.4516
- Rouge1: 20.33
- Rouge2: 6.2
- Rougel: 19.9
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: 0.000275
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20
Training results
Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel |
---|---|---|---|---|---|---|
4.0401 | 1.0 | 388 | 2.5481 | 16.31 | 4.7 | 16.1 |
2.9776 | 2.0 | 776 | 2.4442 | 17.25 | 4.93 | 16.93 |
2.7362 | 3.0 | 1164 | 2.4181 | 19.73 | 5.74 | 19.21 |
2.5767 | 4.0 | 1552 | 2.4071 | 19.37 | 5.62 | 18.89 |
2.4466 | 5.0 | 1940 | 2.3560 | 18.98 | 5.94 | 18.55 |
2.3402 | 6.0 | 2328 | 2.3923 | 20.45 | 5.5 | 20.03 |
2.2385 | 7.0 | 2716 | 2.3639 | 20.03 | 5.96 | 19.76 |
2.1663 | 8.0 | 3104 | 2.3431 | 19.17 | 5.34 | 18.84 |
2.0849 | 9.0 | 3492 | 2.4008 | 19.97 | 5.58 | 19.67 |
2.0203 | 10.0 | 3880 | 2.3948 | 19.67 | 5.75 | 19.26 |
1.9653 | 11.0 | 4268 | 2.3915 | 20.06 | 6.07 | 19.61 |
1.9067 | 12.0 | 4656 | 2.4025 | 20.83 | 6.46 | 20.41 |
1.8592 | 13.0 | 5044 | 2.4194 | 19.97 | 6.4 | 19.69 |
1.8158 | 14.0 | 5432 | 2.4156 | 19.87 | 6.16 | 19.38 |
1.7679 | 15.0 | 5820 | 2.4053 | 19.9 | 5.99 | 19.52 |
1.748 | 16.0 | 6208 | 2.4156 | 19.68 | 5.81 | 19.28 |
1.7198 | 17.0 | 6596 | 2.4306 | 20.0 | 6.26 | 19.63 |
1.6959 | 18.0 | 6984 | 2.4499 | 19.1 | 6.19 | 18.82 |
1.6769 | 19.0 | 7372 | 2.4536 | 20.62 | 6.3 | 20.15 |
1.6682 | 20.0 | 7760 | 2.4516 | 20.33 | 6.2 | 19.9 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1+cu116
- Datasets 2.9.0
- Tokenizers 0.13.2