I use an input dimension of 300 300 when generating the graph so I'm quite confused as to how the expected value is double of the tensor length. I have tried using various shaves to no avail. Please how do I fix this?
There is a separate forum, Intel® Neural Compute Sticks, for these types of queries (start here: https://forums.intel.com/s/topic/0TO0P000000PqZDWA0/intel-neural-compute-sticks?language=en_US).
Unfortunately, I don't have the rights necessary to move your query, so you will need to open a new query there yourself.
Thank you for reaching out.
Please answer the following questions:
- Which version of OpenVINO™ toolkit are you using?
- Which Operative System do you have?
- Which command did you use to convert the model?
- Which sample did you use to test the converted model?
- Which input image or file did you use to test the model?
- Are you using a custom-trained model or a pre-trained model?
- Could you please clarify if you are using an Intel® Movidius™ Neural Compute Stick or an Intel® Neural Compute Stick 2?
Also, can you share with us the mode so we can attempt to convert? You can share it via PM in case you do not want to share it publicly.
Intel Customer Support Technician
A Contingent Worker at Intel