|
--- |
|
license: apache-2.0 |
|
datasets: |
|
- MEDIQA-Chat-2023-taskA |
|
metrics: |
|
- rouge |
|
model-index: |
|
- name: BioBart_Large_Dialouge_Summarization_taskA |
|
results: |
|
- task: |
|
name: Sequence-to-sequence Language Modeling |
|
type: text2text-generation |
|
dataset: |
|
name: MEDIQA-Chat-2023-taskA |
|
type: MEDIQA-Chat-2023-taskA |
|
config: MEDIQA-Chat-2023-taskA |
|
split: train |
|
args: MEDIQA-Chat-2023-taskA |
|
metrics: |
|
- name: Rouge1 |
|
type: rouge |
|
value: 28.7953 |
|
--- |
|
|
|
|
|
|
|
|
|
# BioBart_Large_Dialouge_Summarization_taskA |
|
|
|
This model is a fine-tuned version of [GanjinZero/biobart-large](https://https://huggingface.co/GanjinZero/biobart-large) on the [Task A-Short Dialogue2Note Summarization dataset](https://github.com/abachaa/MEDIQA-Chat-2023) |
|
|
|
It achieves the following results on the evaluation set: |
|
- Loss: 2.7487 |
|
- Rouge1: 28.7953 |
|
- Rouge2: 13.7224 |
|
- Rougel: 27.8491 |
|
- Rougelsum: 28.6028 |
|
- Gen Len: 19.34 |
|
|
|
## 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: 8 |
|
- eval_batch_size: 8 |
|
- num_epochs: 3 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len | |
|
|:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:| |
|
| No log | 1.0 | 1842 | 2.739522 | 29.7742 | 15.4000 | 29.0754 | 29.706200 | 19.5600 | |
|
| No log | 2.0 | 3684 | 2.705775 | 28.6469 | 13.7595 | 27.7582 | 28.486700 | 19.3700 | |
|
| No log | 3.0 | 5526 | 2.748785 | 28.7953 | 13.7224 | 27.8491 | 28.602800 | 19.3400 | |
|
|
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.26.1 |
|
- datasets 2.10.1 |
|
- tokenizers 0.13.2 |