<!-- 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-17jan-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: 2.6637
- Rouge1: 8.3942
- Rouge2: 0.8333
- Rougel: 8.2847
- Rougelsum: 8.3183
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: 0.0001
- train_batch_size: 10
- eval_batch_size: 10
- seed: 42
- 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 | Rouge1 | Rouge2 | Rougel | Rougelsum |
---|---|---|---|---|---|---|---|
11.5311 | 1.0 | 60 | 3.3693 | 3.5755 | 0.6 | 3.6 | 3.5118 |
4.9804 | 2.0 | 120 | 2.9852 | 5.1928 | 0.9667 | 5.205 | 5.1941 |
4.0171 | 3.0 | 180 | 2.8622 | 5.8468 | 0.5889 | 5.9029 | 5.8766 |
3.7179 | 4.0 | 240 | 2.7056 | 8.4114 | 0.5 | 8.5056 | 8.4553 |
3.514 | 5.0 | 300 | 2.7171 | 9.3353 | 0.8333 | 9.2709 | 9.3029 |
3.4154 | 6.0 | 360 | 2.7082 | 8.6179 | 0.4167 | 8.5622 | 8.5483 |
3.3356 | 7.0 | 420 | 2.6801 | 8.3942 | 0.8333 | 8.2847 | 8.3183 |
3.3008 | 8.0 | 480 | 2.6757 | 8.2384 | 0.4167 | 8.1169 | 8.1087 |
3.2493 | 9.0 | 540 | 2.6646 | 8.2384 | 0.4167 | 8.1169 | 8.1087 |
3.2307 | 10.0 | 600 | 2.6637 | 8.3942 | 0.8333 | 8.2847 | 8.3183 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.13.0+cu116
- Datasets 2.8.0
- Tokenizers 0.13.2