Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add files via upload #3

Closed

Conversation

serkansulun
Copy link

@serkansulun serkansulun commented Jul 1, 2018

  • Normalization option to avoid NaNs.
  • Matched author's original implementation with 1% error.
  • Pytorch 4.
  • Manual/adaptive value range.
  • Loops over multiple channels and averages results.
  • Shortened.

- Normalization option to avoid NaNs.
- Matched author's original implementation with 1% error.
- Manual/adaptive value range.
- Loops over multiple channels and averages results.
- Shortened.
@jorge-pessoa
Copy link
Owner

Thank you very much for your contribution ssulun, you caught a few errors/differences between this implementation and the original matlab one which were skewing the results. I have integrated your contributions but not directly from this pull-request as you placed your changes on the incorrect file and you erased a few things I would like to keep due to being compatible with previous code using this package. Additionally it seems your separated the ssim calcuation per channel but as far as I am aware the result is exactly the same as having a multi-channel window. Since it is slightly slower (doesn't use convolution filter optimizations) I have also reverted those changes.

Your contributions and some changes were moved to the branch pr-fixes-ssulun which is now merged into the dev branch. Please check the dev branch and correct me if I got something wrong. And again, thank you very much for your contribution.

@jorge-pessoa
Copy link
Owner

Additionally, I will let this PR open to discuss the integration of your Pull-Request into the dev branch. If you think the changes were correctly merged please let me know or close this Pull-Request.

Thank you,
Jorge

@serkansulun
Copy link
Author

Thanks for the code and everything.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

2 participants