<!-- 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-cnn-pubmed-arxiv-pubmed-v3-e4
This model is a fine-tuned version of theojolliffe/bart-cnn-pubmed-arxiv-pubmed on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.7948
- Rouge1: 52.8917
- Rouge2: 33.9404
- Rougel: 37.0138
- Rougelsum: 50.2918
- Gen Len: 142.0
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: 4
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
---|---|---|---|---|---|---|---|---|
No log | 1.0 | 398 | 0.9591 | 52.9984 | 33.2737 | 34.5312 | 50.3676 | 142.0 |
1.1253 | 2.0 | 796 | 0.8372 | 54.1354 | 34.9653 | 37.381 | 51.0988 | 142.0 |
0.6899 | 3.0 | 1194 | 0.7997 | 52.884 | 34.0614 | 37.6308 | 50.222 | 141.6296 |
0.4982 | 4.0 | 1592 | 0.7948 | 52.8917 | 33.9404 | 37.0138 | 50.2918 | 142.0 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.11.0+cu113
- Datasets 2.1.0
- Tokenizers 0.12.1