[BUG]: represent accidentally converting back to rgb

[logasja]

Before You Report a Bug, Please Confirm You Have Done The Following...

• [X] I have updated to the latest version of the packages.
• [X] I have searched for both existing issues and closed issues and found none that matched my issue.

DeepFace's version

0.0.90

Python version

DeepFace Docker Environment

Operating System

DeepFace Docker Environment

Dependencies

Using DeepFace Docker Environment

Reproducible example

In representation.py starting at line 98:

If skipping detection and loading an image directly, line 82:

img, _ = image_utils.load_image(img_path)

will return a bgr image which is then reversed to rgb at line 103:
```python
img = img[:, :, ::-1]

### Relevant Log Output

_No response_

### Expected Result

_No response_

### What happened instead?

_No response_

### Additional Info

_No response_

[serengil]

It is independent you skip or not detection. Images are fed as rgb to cnn models. Its corresponding comment has typo but it is not a bug.

[logasja]

So then the expected input to the trained models is RGB and not BGR?

[serengil]

Closing this because it is not an issue. But feel free to continue to discuss in this thread if you have any questions or concerns.

[logasja]

Thank you for the clarification, I have been using the models directly and thought they were trained in BGR.