t5-small-finetuned-aspectExtract
This model is a fine-tuned version of t5-small on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.3266
- Rouge1: 66.8064
- Rouge2: 41.6459
- Rougel: 66.027
- Rougelsum: 66.0431
- Gen Len: 3.7994
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: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
---|---|---|---|---|---|---|---|---|
1.823 | 1.0 | 1583 | 1.5166 | 65.1022 | 38.5519 | 64.2732 | 64.2982 | 3.7113 |
1.5931 | 2.0 | 3166 | 1.3623 | 66.4726 | 41.1602 | 65.6868 | 65.6907 | 3.7859 |
1.5285 | 3.0 | 4749 | 1.3266 | 66.8064 | 41.6459 | 66.027 | 66.0431 | 3.7994 |
Framework versions
- Transformers 4.39.0
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 13