mBart50_large_torch
This model is a fine-tuned version of facebook/mbart-large-50 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 3.1233
- Rouge1: 0.257
- Rouge2: 0.0718
- Rougel: 0.187
- Rougelsum: 0.1861
- Gen Len: 88.5902
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: 4
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 140 | 2.6612 | 0.0155 | 0.0045 | 0.0135 | 0.0133 | 183.7541 |
| No log | 2.0 | 280 | 2.6487 | 0.1889 | 0.0497 | 0.1514 | 0.1503 | 88.9508 |
| No log | 3.0 | 420 | 2.8296 | 0.2366 | 0.0693 | 0.1772 | 0.1767 | 84.7705 |
| 1.789 | 4.0 | 560 | 3.1233 | 0.257 | 0.0718 | 0.187 | 0.1861 | 88.5902 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3