<!-- 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-base-finetuned-qg-hard-medium
This model is a fine-tuned version of t5-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.4711
- Rouge1: 44.656
- Rouge2: 24.9885
- Rougel: 40.9697
- Rougelsum: 41.1529
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: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 15
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 135 | 1.5611 | 37.7779 | 19.4817 | 34.3244 | 34.3904 |
| No log | 2.0 | 270 | 1.4731 | 41.8894 | 21.8733 | 37.6817 | 37.6942 |
| No log | 3.0 | 405 | 1.4540 | 43.9334 | 24.786 | 40.4115 | 40.3838 |
| 1.7433 | 4.0 | 540 | 1.4363 | 45.9178 | 26.5837 | 41.7405 | 41.8215 |
| 1.7433 | 5.0 | 675 | 1.4388 | 46.23 | 25.1996 | 41.701 | 41.7289 |
| 1.7433 | 6.0 | 810 | 1.4382 | 46.235 | 25.9074 | 42.3053 | 42.4358 |
| 1.7433 | 7.0 | 945 | 1.4447 | 45.9743 | 26.4922 | 42.107 | 42.244 |
| 1.2283 | 8.0 | 1080 | 1.4490 | 44.3634 | 24.1351 | 40.1315 | 40.2471 |
| 1.2283 | 9.0 | 1215 | 1.4501 | 43.2451 | 23.3871 | 39.7387 | 39.9479 |
| 1.2283 | 10.0 | 1350 | 1.4628 | 44.9832 | 25.2642 | 41.1644 | 41.3158 |
| 1.2283 | 11.0 | 1485 | 1.4621 | 45.6738 | 25.344 | 41.6082 | 41.7572 |
| 1.0817 | 12.0 | 1620 | 1.4667 | 44.6365 | 24.9578 | 40.3016 | 40.4266 |
| 1.0817 | 13.0 | 1755 | 1.4678 | 42.7493 | 22.95 | 38.66 | 38.7194 |
| 1.0817 | 14.0 | 1890 | 1.4708 | 45.2846 | 25.0189 | 41.1739 | 41.3332 |
| 0.9889 | 15.0 | 2025 | 1.4711 | 44.656 | 24.9885 | 40.9697 | 41.1529 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1+cu113
- Datasets 2.7.1
- Tokenizers 0.13.2
