<!-- 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-base-finetuned-cnn_dailymail
This model is a fine-tuned version of facebook/bart-base on the cnn_dailymail dataset. It achieves the following results on the evaluation set:
- Loss: 1.5396
- Rouge1: 0.3511
- Rouge2: 0.1925
- Rougel: 0.3086
- Rougelsum: 0.3292
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: 5.6e-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: 4
Training results
Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
---|---|---|---|---|---|---|---|
1.9486 | 1.0 | 35890 | 1.5941 | 0.3498 | 0.1893 | 0.3063 | 0.3272 |
1.6706 | 2.0 | 71780 | 1.5601 | 0.3503 | 0.1916 | 0.3079 | 0.3279 |
1.4809 | 3.0 | 107670 | 1.5423 | 0.3520 | 0.1923 | 0.3086 | 0.3295 |
1.3293 | 4.0 | 143560 | 1.5396 | 0.3511 | 0.1925 | 0.3086 | 0.3292 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1+cu113
- Datasets 2.7.1
- Tokenizers 0.13.2