<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
bart-large-cnn-finetuned-weaksup-1000-pad-early-new1
This model is a fine-tuned version of facebook/bart-large-cnn on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4948
- Rouge1: 28.1465
- Rouge2: 13.4076
- Rougel: 22.2763
- Rougelsum: 25.2087
- Gen Len: 68.58
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: 1
- eval_batch_size: 1
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| 0.156 | 1.0 | 1000 | 0.4377 | 27.8782 | 13.1274 | 21.2329 | 24.6465 | 66.25 |
| 0.0843 | 2.0 | 2000 | 0.4948 | 28.1465 | 13.4076 | 22.2763 | 25.2087 | 68.58 |
Framework versions
- Transformers 4.16.2
- Pytorch 1.10.2
- Datasets 1.18.3
- Tokenizers 0.11.0
