<!-- 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. -->
t5-small-finetuned-xsum
This model is a fine-tuned version of t5-small on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0078
- Rouge1: 48.0041
- Rouge2: 41.1108
- Rougel: 48.0417
- Rougelsum: 48.0048
- Gen Len: 19.0
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: 2e-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: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 125 | 0.8794 | 6.6951 | 0.2224 | 5.4261 | 5.4191 | 18.59 |
| No log | 2.0 | 250 | 0.1755 | 41.0386 | 32.7852 | 41.06 | 41.0268 | 19.0 |
| No log | 3.0 | 375 | 0.0702 | 38.8519 | 30.7641 | 38.8569 | 38.8409 | 19.0 |
| 0.9138 | 4.0 | 500 | 0.0436 | 42.2193 | 34.4917 | 42.2304 | 42.2051 | 19.0 |
| 0.9138 | 5.0 | 625 | 0.0315 | 42.069 | 34.3637 | 42.1018 | 42.0801 | 19.0 |
| 0.9138 | 6.0 | 750 | 0.0233 | 45.9645 | 38.6498 | 45.9851 | 45.958 | 19.0 |
| 0.9138 | 7.0 | 875 | 0.0198 | 45.9049 | 38.6707 | 45.9519 | 45.9069 | 19.0 |
| 0.0769 | 8.0 | 1000 | 0.0175 | 45.8588 | 38.5195 | 45.8955 | 45.8691 | 19.0 |
| 0.0769 | 9.0 | 1125 | 0.0148 | 46.487 | 39.3562 | 46.5325 | 46.4933 | 19.0 |
| 0.0769 | 10.0 | 1250 | 0.0131 | 46.8666 | 39.7413 | 46.9032 | 46.8584 | 19.0 |
| 0.0769 | 11.0 | 1375 | 0.0117 | 47.1968 | 40.1561 | 47.2436 | 47.1913 | 19.0 |
| 0.0444 | 12.0 | 1500 | 0.0106 | 47.4808 | 40.4857 | 47.5092 | 47.4752 | 19.0 |
| 0.0444 | 13.0 | 1625 | 0.0098 | 47.4893 | 40.4966 | 47.5301 | 47.5073 | 19.0 |
| 0.0444 | 14.0 | 1750 | 0.0092 | 47.5574 | 40.573 | 47.5867 | 47.5567 | 19.0 |
| 0.0444 | 15.0 | 1875 | 0.0088 | 47.7141 | 40.7683 | 47.753 | 47.7052 | 19.0 |
| 0.0333 | 16.0 | 2000 | 0.0084 | 47.7701 | 40.8465 | 47.8078 | 47.7729 | 19.0 |
| 0.0333 | 17.0 | 2125 | 0.0081 | 47.8671 | 40.9538 | 47.9002 | 47.8609 | 19.0 |
| 0.0333 | 18.0 | 2250 | 0.0079 | 47.8993 | 40.993 | 47.9353 | 47.8973 | 19.0 |
| 0.0333 | 19.0 | 2375 | 0.0079 | 48.01 | 41.1113 | 48.0419 | 48.0021 | 19.0 |
| 0.029 | 20.0 | 2500 | 0.0078 | 48.0041 | 41.1108 | 48.0417 | 48.0048 | 19.0 |
Framework versions
- Transformers 4.33.1
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.13.3
