chore: Add efficientnet b0 and VIT to testsuite

[peri044]

Description

Add efficientnet and vit to testsuite

Type of change

• New feature (non-breaking change which adds functionality)

Checklist:

• My code follows the style guidelines of this project (You can use the linters)
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas and hacks
• I have made corresponding changes to the documentation
• I have added tests to verify my fix or my feature
• New and existing unit tests pass locally with my changes

[github-actions]

Code conforms to C++ style guidelines

[github-actions]

Code conforms to Python style guidelines

[github-actions]

Code conforms to Python style guidelines

[github-actions]

Code conforms to C++ style guidelines

[narendasan]

Can we change these tests slightly so that the threshold is an argument? would be good not to lower the threshold for the existing networks

[peri044]

@narendasan The threshold needs to be increased for VIT. I think it might be because of GeLU. The difference for a test image that I saw is around (6~7)e-3. Accuracy on imagenet seems to be almost similar to pytorch.

[github-actions]

Code conforms to Python style guidelines

[github-actions]

Code conforms to C++ style guidelines

[narendasan]

Yeah so we should parameterize the threshold like we do the model and the input size so that only VIT uses the lower threshold

[github-actions]

Code conforms to Python style guidelines

[github-actions]

Code conforms to C++ style guidelines

[peri044]

Yeah so we should parameterize the threshold like we do the model and the input size so that only VIT uses the lower threshold

Done.

[github-actions]

Code conforms to C++ style guidelines

[github-actions]

Code conforms to Python style guidelines

[narendasan]

lgtm