-
Notifications
You must be signed in to change notification settings - Fork 1.7k
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
MNN does not use GPU mail-G31 for inference on android #3093
Comments
Use RuntimeManager's getInfo interface can get the real backend. |
Otherwise, do you compile MNN with -DMNN_OPENCL=true and use System.load("libMNN_OpenCL") in app's java code? |
Hi @jxt1234
This is the CMakeLists.txt file
If I add System.load("libMNN_OpenCL") to kotlin code as above, it is crashed. This is OpenCL version on my device
This is about my GPU info on device
Plz help me check why MNN does not use GPU to infer |
you can add -DMNN_SEP_BUILD=OFF to build MNN. Then it will only build a libMNN.so |
Hi everyone
I am deploying model yolov11 on Android NDK using the MNN platform. I follow this link and use OpenCL for inference.
Hear is my code
Although using the MNN_FORWARD_OPENCL I see that running it uses a lot of CPU and time inference is the same as CPU . I don't know why.
My device has GPU mail-G31
Thanks everyone !!!
The text was updated successfully, but these errors were encountered: