metadata
license: mit
base_model: facebook/bart-large-xsum
tags:
- generated_from_trainer
metrics:
- rouge
model-index:
- name: LLM_Teach_Bart
results: []
LLM_Teach_Bart
This model is a fine-tuned version of facebook/bart-large-xsum on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.5031
- Rouge1: 0.4867
- Rouge2: 0.2549
- Rougel: 0.376
- Rougelsum: 0.3764
- Gen Len: 46.2273
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: 16
- 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
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| 1.3272 | 1.0 | 625 | 1.3454 | 0.4905 | 0.2577 | 0.3725 | 0.3726 | 51.3691 |
| 0.9972 | 2.0 | 1250 | 1.3446 | 0.4861 | 0.26 | 0.3757 | 0.3761 | 46.5909 |
| 0.8102 | 3.0 | 1875 | 1.4353 | 0.4889 | 0.2564 | 0.3753 | 0.3755 | 47.3073 |
| 0.5636 | 4.0 | 2500 | 1.5031 | 0.4867 | 0.2549 | 0.376 | 0.3764 | 46.2273 |
Framework versions
- Transformers 4.36.0
- Pytorch 2.0.1+cu117
- Datasets 2.14.5
- Tokenizers 0.15.0