long-t5-tglobal-base-mediasum
This model is a fine-tuned version of pszemraj/long-t5-tglobal-base-16384-book-summary on the multi_news dataset. It achieves the following results on the evaluation set:
- Loss: 2.0387
- Rouge1: 0.3246
- Rouge2: 0.0867
- Rougel: 0.1663
- Rougelsum: 0.1662
- Gen Len: 106.985
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
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| 2.4191 | 1.0 | 4500 | 2.0952 | 0.3389 | 0.0882 | 0.1706 | 0.1706 | 118.285 |
| 2.3462 | 2.0 | 9000 | 2.0484 | 0.3339 | 0.0887 | 0.1683 | 0.1683 | 111.936 |
| 2.3458 | 3.0 | 13500 | 2.0387 | 0.3246 | 0.0867 | 0.1663 | 0.1662 | 106.985 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3