distilbart-xsum-12-1-finetuned-pubmed
This model is a fine-tuned version of sshleifer/distilbart-xsum-12-1 on the pub_med_summarization_dataset dataset. It achieves the following results on the evaluation set:
- Loss: 2.8236
- Rouge1: 27.0012
- Rouge2: 12.728
- Rougel: 19.8685
- Rougelsum: 25.0485
- Gen Len: 59.969
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: 2
- eval_batch_size: 2
- 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 |
|---|---|---|---|---|---|---|---|---|
| 3.3604 | 1.0 | 4000 | 3.1575 | 25.0078 | 11.5381 | 18.4246 | 23.1605 | 54.8935 |
| 3.0697 | 2.0 | 8000 | 2.9478 | 26.4947 | 12.5411 | 19.4328 | 24.6123 | 57.948 |
| 2.8638 | 3.0 | 12000 | 2.8672 | 26.8856 | 12.7568 | 19.8949 | 24.8745 | 59.6245 |
| 2.7243 | 4.0 | 16000 | 2.8347 | 26.7347 | 12.5152 | 19.6516 | 24.7756 | 60.439 |
| 2.6072 | 5.0 | 20000 | 2.8236 | 27.0012 | 12.728 | 19.8685 | 25.0485 | 59.969 |
Framework versions
- Transformers 4.17.0
- Pytorch 1.10.0+cu111
- Datasets 1.18.3
- Tokenizers 0.11.6