<!-- 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. -->
flan-t5-small-samsum
This model is a fine-tuned version of google/flan-t5-small on the samsum dataset. It achieves the following results on the evaluation set:
- Loss: 1.6191
- Rouge1: 44.1381
- Rouge2: 20.1928
- Rougel: 36.5369
- Rougelsum: 40.4069
- Gen Len: 16.9377
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: 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: 20.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| 1.8119 | 1.0 | 1842 | 1.6673 | 42.607 | 19.0937 | 35.7568 | 39.0753 | 16.6642 |
| 1.7451 | 2.0 | 3684 | 1.6427 | 43.8834 | 19.669 | 36.4633 | 39.9905 | 17.1258 |
| 1.6756 | 3.0 | 5526 | 1.6333 | 43.3665 | 19.5917 | 36.2074 | 39.6598 | 16.6313 |
| 1.6411 | 4.0 | 7368 | 1.6258 | 44.0229 | 19.7965 | 36.3871 | 40.1807 | 16.9158 |
| 1.5939 | 5.0 | 9210 | 1.6265 | 44.3267 | 19.91 | 36.6279 | 40.5258 | 17.2015 |
| 1.5619 | 6.0 | 11052 | 1.6197 | 43.834 | 19.5188 | 36.2175 | 40.0349 | 17.0 |
| 1.5328 | 7.0 | 12894 | 1.6211 | 44.2587 | 20.0519 | 36.5957 | 40.3573 | 17.0061 |
| 1.528 | 8.0 | 14736 | 1.6191 | 44.1381 | 20.1928 | 36.5369 | 40.4069 | 16.9377 |
| 1.5 | 9.0 | 16578 | 1.6264 | 44.2013 | 20.0435 | 36.6233 | 40.5776 | 16.9866 |
| 1.473 | 10.0 | 18420 | 1.6212 | 44.4443 | 20.2859 | 36.7695 | 40.643 | 16.9512 |
| 1.4765 | 11.0 | 20262 | 1.6227 | 44.5579 | 20.3044 | 37.0447 | 40.8628 | 16.9328 |
| 1.4361 | 12.0 | 22104 | 1.6300 | 44.3329 | 20.543 | 36.7583 | 40.6138 | 16.7131 |
| 1.4296 | 13.0 | 23946 | 1.6285 | 44.7236 | 20.5361 | 37.1228 | 41.0207 | 17.1746 |
| 1.4119 | 14.0 | 25788 | 1.6314 | 44.5903 | 20.7486 | 37.1988 | 40.9799 | 16.8657 |
| 1.4075 | 15.0 | 27630 | 1.6323 | 44.8141 | 20.849 | 37.5116 | 41.2015 | 16.9634 |
| 1.4054 | 16.0 | 29472 | 1.6314 | 44.9286 | 20.8482 | 37.5008 | 41.1902 | 16.9072 |
| 1.3984 | 17.0 | 31314 | 1.6337 | 44.8266 | 20.8771 | 37.2837 | 41.2143 | 17.0 |
| 1.3862 | 18.0 | 33156 | 1.6344 | 44.9681 | 20.7656 | 37.4363 | 41.2447 | 16.9707 |
| 1.3679 | 19.0 | 34998 | 1.6346 | 44.8412 | 20.6233 | 37.2335 | 41.1223 | 16.9512 |
| 1.3702 | 20.0 | 36840 | 1.6346 | 44.5877 | 20.5694 | 37.093 | 40.9197 | 16.9414 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0
- Datasets 2.10.1
- Tokenizers 0.13.2
