metadata
license: mit
base_model: facebook/bart-large-xsum
tags:
- generated_from_trainer
metrics:
- rouge
model-index:
- name: bart_samsum
results: []
bart_samsum
This model is a fine-tuned version of facebook/bart-large-xsum on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.6098
- Rouge1: 52.3831
- Rouge2: 27.5513
- Rougel: 43.5051
- Rougelsum: 48.1509
- Gen Len: 30.1941
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: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- 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 |
|---|---|---|---|---|---|---|---|---|
| 1.4242 | 0.9997 | 1841 | 1.5158 | 52.6686 | 27.364 | 43.1196 | 47.9363 | 30.5824 |
| 1.0951 | 2.0 | 3683 | 1.5060 | 52.8177 | 27.6542 | 43.6251 | 48.207 | 30.2051 |
| 0.8624 | 2.9997 | 5524 | 1.5495 | 52.6928 | 28.1014 | 43.8451 | 48.4256 | 28.4212 |
| 0.6984 | 3.9989 | 7364 | 1.6098 | 52.3831 | 27.5513 | 43.5051 | 48.1509 | 30.1941 |
Framework versions
- Transformers 4.40.0
- Pytorch 2.2.1+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1