[Fix] MMBench Doc Fix (#96)

* update

* update

* fix lint
This commit is contained in:
Haodong Duan 2023-07-25 10:43:22 +08:00 committed by GitHub
parent 3715be6595
commit 4b0aa80466
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

View File

@ -87,7 +87,7 @@ class MMBenchDataset(Dataset):
## How to construct the inference prompt
```python
if data_sample['context'] is None:
if data_sample['context'] is not None:
prompt = data_sample['context'] + ' ' + data_sample['question'] + ' ' + data_sample['options']
else:
prompt = data_sample['question'] + ' ' + data_sample['options']
@ -101,11 +101,15 @@ C. Digital art
D. Photo
<div align=center>
<img src="https://user-images.githubusercontent.com/56866854/252847545-ea829a95-b063-492f-8760-d27143b5c834.jpg" width="10%"/>
<img src="https://github-production-user-asset-6210df.s3.amazonaws.com/34324155/255581681-1364ef43-bd27-4eb5-b9e5-241327b1f920.png" width="50%"/>
</div>
```
prompt = ###Human: Question: Which category does this image belong to? There are several options: A. Oil Painting, B. Sketch, C. Digital art, D. Photo ###Assistant:
```python
prompt = """
###Human: Question: Which category does this image belong to?
There are several options: A. Oil Painting, B. Sketch, C. Digital art, D. Photo
###Assistant:
"""
```
You can make custom modifications to the prompt