t5_small_samsum
This model is a fine-tuned version of google-t5/t5-small on the samsum dataset. It achieves the following results on the evaluation set:
- Loss: 1.7255
- Rouge1: 0.4282
- Rouge2: 0.2003
- Rougel: 0.36
- Rougelsum: 0.3596
- Gen Len: 16.7372
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: 3e-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 |
|---|---|---|---|---|---|---|---|---|
| 1.9452 | 1.0 | 921 | 1.7726 | 0.4147 | 0.1901 | 0.3492 | 0.3493 | 16.4719 |
| 1.8952 | 2.0 | 1842 | 1.7498 | 0.4237 | 0.1971 | 0.3577 | 0.3577 | 16.4548 |
| 1.8703 | 3.0 | 2763 | 1.7323 | 0.4243 | 0.1968 | 0.3571 | 0.3566 | 16.7689 |
| 1.8579 | 4.0 | 3684 | 1.7310 | 0.4262 | 0.2012 | 0.3606 | 0.3604 | 16.7641 |
| 1.8525 | 5.0 | 4605 | 1.7255 | 0.4282 | 0.2003 | 0.36 | 0.3596 | 16.7372 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.1
- Tokenizers 0.19.1