fixed super-gradient installation on yolo-nas notebook

[arrafi-musabbir]

Description

The current method !pip install -q super-gradients==3.1.1 to install super-gradient from pip results in pycocotools build failure. I followed the deci-ai's repository and found the proper way to install it. The following command will result in a successful installation:
!pip install git+https://github.com/Deci-AI/super-gradients.git@stable

Type of change

Please delete options that are not relevant.

• Bug fix (non-breaking change which fixes an issue)

How has this change been tested, please provide a testcase or example of how you tested the change?

I just changed the way super-gradient package was being installed previously
!pip install git+https://github.com/Deci-AI/super-gradients.git@stable

Any specific deployment considerations

For example, documentation changes, usability, usage/costs, secrets, etc.

• n/a

Docs

• Docs updated? What were the changes:

In the Install YOLO-NAS block I fixed the way super-gradient is supposed to be installed correctly

I replaced this line !pip install -q super-gradients==3.1.1 with this: !pip install git+https://github.com/Deci-AI/super-gradients.git@stable

Now the pycocotools build failure is resolved.

[CLAassistant]

All committers have signed the CLA.