metadata
base_model: facebook/bart-base
library_name: peft
license: apache-2.0
metrics:
- rouge
tags:
- generated_from_trainer
model-index:
- name: bart-base-summarization-medical-46
results: []
bart-base-summarization-medical-46
This model is a fine-tuned version of facebook/bart-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.1274
- Rouge1: 0.4211
- Rouge2: 0.2237
- Rougel: 0.3559
- Rougelsum: 0.356
- Gen Len: 18.296
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: 3e-05
- train_batch_size: 4
- eval_batch_size: 1
- seed: 46
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| 2.7041 | 1.0 | 1250 | 2.2079 | 0.4109 | 0.2179 | 0.35 | 0.3501 | 18.052 |
| 2.6134 | 2.0 | 2500 | 2.1641 | 0.4151 | 0.2203 | 0.3517 | 0.3518 | 18.103 |
| 2.5664 | 3.0 | 3750 | 2.1450 | 0.4163 | 0.22 | 0.353 | 0.3529 | 18.051 |
| 2.5501 | 4.0 | 5000 | 2.1328 | 0.4191 | 0.2217 | 0.353 | 0.3528 | 18.228 |
| 2.535 | 5.0 | 6250 | 2.1268 | 0.4221 | 0.2253 | 0.3569 | 0.3568 | 18.235 |
| 2.5102 | 6.0 | 7500 | 2.1274 | 0.4211 | 0.2237 | 0.3559 | 0.356 | 18.296 |
Framework versions
- PEFT 0.12.0
- Transformers 4.42.4
- Pytorch 2.3.1+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1