We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
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
Some ideas, on top of my head.
We are using Simd to accelerate some core image processing functions (e.g. color conversion, image resizing, ...) because:
Simd
x86
ARM
What would be great:
CMake
AVX-512
color conversion
image resizing
src Simd Color conversion ARM x86 [...] Image resize ARM x86 [...] [...]
The text was updated successfully, but these errors were encountered:
Hi! Thank you for your suggestions.
I want to note that there are several cmake options in prj/cmake/CMakeLists.txt which can reduce compilation time:
prj/cmake/CMakeLists.txt
option(SIMD_AVX512 "AVX-512 (AVX-512F, AVX-512CD, AVX-512VL, AVX-512DQ, AVX-512BW) enable" ON) option(SIMD_AVX512VNNI "AVX-512-VNNI enable" ON) option(SIMD_TEST "Test framework enable" ON) option(SIMD_SYNET "Synet optimizations enable" ON) option(SIMD_INT8_DEBUG "Synet INT8 debug enable" OFF)
I would recommend to switch off them and you will see radical reducing of compilation time!
As to changing of project structure I have simular plans but them have low prority.
Sorry, something went wrong.
No branches or pull requests
Some ideas, on top of my head.
Simd modules
We are using
Simd
to accelerate some core image processing functions (e.g. color conversion, image resizing, ...) because:x86
andARM
, the architectures we are considering.What would be great:
Simd
lib:CMake
(what we are using),AVX-512
(I believe is still marginal for regular CPU and reduces the CPU frequency to avoid overheating?)color conversion
,image resizing
, ... to avoid long build time and big library sizeOther ideas
The text was updated successfully, but these errors were encountered: