I have installed openvino2021 and succefully worked out till at the step of loading model to device i am having error for driver for movidius so i was finiding it in intel files but it is not there where can i find that .info file?
Thank you for posting on the Intel® communities.
We would like to inform you that we have a forum for these specific issues and products, so we are moving it to the "Software Development SDKs and Libraries > Intel® Distribution of OpenVINO™ Toolkit" forum so you can get better support for this matter.
Intel Customer Support Technician
First and foremost, please note that With OpenVINO™ 2020.4 release, Intel® Movidius™ Neural Compute Stick is no longer supported. You might want to consider migrating to Neural COmpute Stick2 (NCS2).
You may refer here for the official guide: https://docs.openvinotoolkit.org/latest/openvino_docs_install_guides_installing_openvino_windows.htm....
For inferencing with NCS2 (-d MYRIAD) you need to follow some additional steps to install the driver.
You may refer to this official guide: https://docs.openvinotoolkit.org/latest/openvino_docs_install_guides_installing_openvino_windows.htm...
thanks alot for your kind reply just another query here : i am working over unet model ,one that is provided over github
and second one
I want to ask that they both are supported by NCS 2 or not?
I have IR files for both of these models.
how could i use these IR files with NCS 2 i am unaware of the last step if you can guide me here it means alot for me
thanks in advance
If the model is supported by OpenVINO, then running on the NCS2 shouldn't be a problem.
However, so far OpenVINO only supports the models of Tensorflow, ONNX, Caffe, MXNet, Kaldi.
You may refer to this guide for more info including how to convert them. The topologies supported for each of those models are also listed here: https://docs.openvinotoolkit.org/latest/openvino_docs_MO_DG_prepare_model_convert_model_Convert_Mode...
Intel will no longer monitor this thread since we have provided a solution. If you need any additional information from Intel, please submit a new question.