-
Notifications
You must be signed in to change notification settings - Fork 57
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
Caffe to NNEF: No support for Prelu's #72
Comments
Indeed, this is missing from the converter, we'll add it soon. |
We have added prelu support, can you test it now? |
Yes the model now successfully converts thanks! |
It believe there was no official such converter. Where was that converter available? |
This was what I was thinking of, but its AMD specific:
https://github.com/GPUOpen-ProfessionalCompute-Libraries/amdovx-modules/blob/develop/utils/model_compiler/README.md
Any plans for a more generic one ?
…________________________________
From: Viktor Gyenes <notifications@github.com>
Sent: 02 April 2019 16:23:36
To: KhronosGroup/NNEF-Tools
Cc: Neil Jones; Author
Subject: Re: [KhronosGroup/NNEF-Tools] Caffe to NNEF: No support for Prelu's (#72)
It believe there was no official such converter. Where was that converter available?
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub<#72 (comment)>, or mute the thread<https://github.com/notifications/unsubscribe-auth/Aui6IZsc9Q001DvUEXMBn115NMwFb_R9ks5vc3X4gaJpZM4cCKDu>.
|
I believe the OpenVX group is working on a mechanism to import NNEF models into OpenVX, that can be implemented by vendors, but I cannot tell details about it until the spec is published. |
I'm trying to convert MTCNN to NNEF (and ultimately to openVX) but the Caffe converter doesn't support Prelu activations:
~/src/MTCNN_face_detection_alignment/code/codes/MTCNNv1/model$ caffe_to_nnef det1.prototxt --model det1.caffemodel
Error: caffe_to_dog: No converter for: PReLU
Error: caffe_to_dog: No converter for: PReLU
Error: caffe_to_dog: No converter for: PReLU
Error: No converter for PReLU
Error: conv1_1: Corresponding targetdn not defined
Error: No converter for PReLU
Error: conv2_1: Corresponding targetdn not defined
Error: No converter for PReLU
Error: conv3_1: Corresponding targetdn not defined
https://github.com/kpzhang93/MTCNN_face_detection_alignment
The text was updated successfully, but these errors were encountered: