How can I use importONNXNetwork function on a production machine which only has MCR installed

13 次查看(过去 30 天)
Hi,
I want to use a pre-trained ONNX neural network on a production machine (PC) which only has MCR no Matlab. I tested my code in my development machine with MATLAB Version: 9.7.0.1190202 (R2019b), there was no problem. But once I complied it into a .net assembly and tried to use it on the production machine, it threw an error:
importONNXNetwork: importONNXNetwork requires the Deep Learning Toolbox Converter for ONNX Model Format support package. To install this support package, use the <a href="matlab: matlab.addons.supportpackage.internal.explorer.showSupportPackages('ONNXCONVERTER', 'tripwire')">Add-On Explorer</a>.
I think the problem is that the add-on "Deep Learning Toolbox Converter for ONNX Model" is not installed on the production machine. In contrast, my code can run without issue on my development machine because "Deep Learning Toolbox Converter for ONNX Model" is installed there.
How can I install the add-on "Deep Learning Toolbox Converter for ONNX Model" on the production machine with MCR? Or is there any other way to use a pre-trained ONNX or other types of neural network on a production machine (PC) which only has MCR no Matlab?
Thank you!

采纳的回答

CHANGYUAN
CHANGYUAN 2020-1-22
Thanks to the support from Sandesh Joshi at MathWorks Technical Support Department. It worked when I followed his instruction.
Add the following in the "Additional parameters passed to mcc" section in the settings page of library compiler:
-a C:\ProgramData\MATLAB\SupportPackages\R2019b\toolbox\nnet\supportpackages\onnx -a C:\ProgramData\MATLAB\SupportPackages\R2019b\resources\nnet_cnn_onnx

更多回答(0 个)

类别

Help CenterFile Exchange 中查找有关 Image Data Workflows 的更多信息

产品


版本

R2019b

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by