<!-- 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. -->
tl_dr_summary
This model is a fine-tuned version of t5-small on the tldr_news dataset. It achieves the following results on the evaluation set:
- Loss: 2.6215
- Rouge1: 0.408
- Rouge2: 0.2294
- Rougel: 0.3802
- Rougelsum: 0.3805
- Gen Len: 14.2469
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: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 447 | 2.7918 | 0.3234 | 0.1352 | 0.2944 | 0.2941 | 15.9559 |
| 3.4426 | 2.0 | 894 | 2.6821 | 0.398 | 0.2216 | 0.3704 | 0.3704 | 14.4849 |
| 3.0161 | 3.0 | 1341 | 2.6415 | 0.4042 | 0.2266 | 0.3765 | 0.3765 | 14.4106 |
| 2.9296 | 4.0 | 1788 | 2.6272 | 0.4068 | 0.2292 | 0.3796 | 0.3802 | 14.228 |
| 2.8791 | 5.0 | 2235 | 2.6215 | 0.408 | 0.2294 | 0.3802 | 0.3805 | 14.2469 |
Framework versions
- Transformers 4.27.4
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3
