metadata
base_model: ubaada/lsg-bart-large-4096-booksum
tags:
- generated_from_trainer
metrics:
- rouge
model-index:
- name: lsg-bart-large-4096-booksum
results: []
lsg-bart-large-4096-booksum
This model is a fine-tuned version of ubaada/lsg-bart-large-4096-booksum on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.0742
- Rouge1: 0.4145
- Rouge2: 0.0797
- Rougel: 0.1541
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: 8e-05
- train_batch_size: 8
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel |
|---|---|---|---|---|---|---|
| 1.3801 | 1.0 | 1251 | 2.0441 | 0.4223 | 0.0811 | 0.1532 |
| 1.2385 | 2.0 | 2502 | 2.0753 | 0.3995 | 0.0751 | 0.1512 |
| 0.9542 | 3.0 | 3753 | 2.0742 | 0.4145 | 0.0797 | 0.1541 |
Framework versions
- Transformers 4.40.2
- Pytorch 2.2.0
- Datasets 2.19.1
- Tokenizers 0.19.1