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

How to implement a simple 2D convolution (FFT + point_wise_product + iFFT) #193

Open
jasmineforever opened this issue Nov 28, 2024 · 1 comment

Comments

@jasmineforever
Copy link

Hello:
I'm new to vkfft, it's awesome! I still learning how to use it. I read the sample codes and didn't see a similar pattern like “FFT + point_wise_product + iFFT” convolution. Does vkfft support such a pattern?

@DTolm
Copy link
Owner

DTolm commented Dec 3, 2024

Hello,

Sorry for the slow reply, for convolutions you can see samples 50-52 - they all do this pattern. You can also check section 5.6 of the documentation. Essentially you need to create the kernel with separate kernel application with the following parameter
kernel_configuration.kernelConvolution = 1;
and then create main application with
convolution_configuration.performConvolution = 1;
and pass kernel in convolution_configuration.kernel.

Hope this works for you and if not I will happily try to answer your questions.

Best regards,
Dmitrii

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

No branches or pull requests

2 participants