#161 update macs calibration library

[deepkyu]

Description

Please include a summary in English, of the changes in this pull request. If it closes an issue, please mention it here.

Closes: #161

You should link at least one existing issue for PR. Before your create a PR, please check to see if there is an issue for this change.
PRs from forked repository not accepted.

Change(s)

• Synchronize MACs calibration library with PyNetsPresso and NetsPresso platform
  • https://docs.netspresso.ai/docs/release-notes

Changelog

If you PR to dev branch, please add a brief summary of the change to the Upcoming Release section of the CHANGELOG.md file and include a link to the PR (formatted in markdown) and a link to your github profile.

For example,

- Added a new feature by `@myusername` in [PR 2023](https://github.com/Nota-NetsPresso/netspresso-trainer/pull/2023)

Code Formatting

If you PR to either master or dev branch, you should follow the code linting process. Please check your code with lint_check.sh in ./scripts directory.
For more information, please read the contribution guide in CONTRIBUTING.md.

[deepkyu]

Added special reviewer @Only-bottle for:

• Validate the counter update

[deepkyu]

fvcore does not separately count pooling layers and add_ operators. This issue is also mentioned in the fvcore issue.

In addition, fvcore does not provide a separate count for params, so I have changed it to the commonly used one, params = sum(p.numel() for p in model.parameters()). This is equivalent with the caliration procedure in thop.
As a result, in NP Trainer ResNet50, I have confirmed the following results. Note that the params remain the same, but there is a difference in MACs.

fvcore + vanila params count

2023-10-23_06:47:31 UTC | INFO          | save_summary:<base.py>:322 >>> [Model stats] Params: 23.51M | MACs: 5.37G

thop

2023-10-23_06:48:32 UTC | INFO          | save_summary:<base.py>:322 >>> [Model stats] Params: 23.51M | MACs: 5.40G

[Only-bottle]

👍🏻

[illian01]

LGTM!