<!-- 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-med-term-conditional-masking
This model is a fine-tuned version of facebook/bart-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5115
- Rouge2 Precision: 0.7409
- Rouge2 Recall: 0.5343
- Rouge2 Fmeasure: 0.6025
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: 8
- eval_batch_size: 8
- 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 | Rouge2 Precision | Rouge2 Recall | Rouge2 Fmeasure |
---|---|---|---|---|---|---|
0.6278 | 1.0 | 15827 | 0.5546 | 0.7255 | 0.5244 | 0.5908 |
0.5356 | 2.0 | 31654 | 0.5286 | 0.7333 | 0.5293 | 0.5966 |
0.4757 | 3.0 | 47481 | 0.5154 | 0.7376 | 0.532 | 0.5998 |
0.4337 | 4.0 | 63308 | 0.5107 | 0.7406 | 0.5342 | 0.6023 |
0.4045 | 5.0 | 79135 | 0.5115 | 0.7409 | 0.5343 | 0.6025 |
Framework versions
- Transformers 4.17.0
- Pytorch 1.10.0+cu111
- Datasets 2.0.0
- Tokenizers 0.11.6