<!-- 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-model2-3110-e4
This model is a fine-tuned version of theojolliffe/bart-paraphrase-v4-e1-feedback on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0700
- Rouge1: 70.0692
- Rouge2: 68.1457
- Rougel: 69.8943
- Rougelsum: 70.0389
- Gen Len: 19.8966
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: 6
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
---|---|---|---|---|---|---|---|---|
0.5951 | 1.0 | 553 | 0.3089 | 62.5675 | 54.7411 | 61.2646 | 61.3675 | 19.7241 |
0.2541 | 2.0 | 1106 | 0.1432 | 66.113 | 61.964 | 64.6141 | 64.9187 | 19.8966 |
0.1547 | 3.0 | 1659 | 0.0964 | 68.6902 | 64.938 | 67.6197 | 67.9181 | 19.8966 |
0.1141 | 4.0 | 2212 | 0.1015 | 68.9122 | 66.4279 | 68.4906 | 68.5758 | 19.8966 |
0.0728 | 5.0 | 2765 | 0.0819 | 69.2271 | 66.8276 | 68.6915 | 68.849 | 19.8966 |
0.0563 | 6.0 | 3318 | 0.0700 | 70.0692 | 68.1457 | 69.8943 | 70.0389 | 19.8966 |
Framework versions
- Transformers 4.23.1
- Pytorch 1.12.1+cu113
- Datasets 2.6.1
- Tokenizers 0.13.1