t5-small-finetuned-cc-news-es-titles
This model is a fine-tuned version of t5-small on the cc-news-es-titles dataset. It achieves the following results on the evaluation set:
- Loss: 2.6383
- Rouge1: 16.701
- Rouge2: 4.1265
- Rougel: 14.8175
- Rougelsum: 14.8193
- Gen Len: 18.9159
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: 1
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| 2.8439 | 1.0 | 23133 | 2.6383 | 16.701 | 4.1265 | 14.8175 | 14.8193 | 18.9159 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.12.0+cu113
- Datasets 2.3.2
- Tokenizers 0.12.1