<!-- 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. -->
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: 2.9535
- Rouge1: 0.1566
- Rouge2: 0.0333
- Rougel: 0.1242
- Rougelsum: 0.124
- Gen Len: 88.63
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 |
|---|---|---|---|---|---|---|---|---|
| 3.1667 | 1.0 | 800 | 2.6328 | 0.102 | 0.0214 | 0.091 | 0.0909 | 81.98 |
| 1.9765 | 2.0 | 1600 | 2.5806 | 0.1317 | 0.0273 | 0.1096 | 0.1099 | 87.67 |
| 1.4771 | 3.0 | 2400 | 2.7266 | 0.1482 | 0.0321 | 0.1225 | 0.1227 | 92.88 |
| 1.0086 | 4.0 | 3200 | 2.9535 | 0.1566 | 0.0333 | 0.1242 | 0.124 | 88.63 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
