- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi there!
So, my company just bought a Neural Compute Stick. Our plan is to develop some nice deep learning models, deploy them to the NCS, connect it to a raspeberry pi 3 (with a camera also connected) and run some fun experiments. Maybe recognizing some faces, or palying around with a toy self driving car, we still don't know. Well, it is my job now to learn about deep learning and understand how to work with the NCS.
And that's the problem. In my computer (running Windows), I've been playing with keras and tensorflow and was able to develop some simple networks. What I would like to do now is to train a model in my computer, save it on the NCS and running some examples on it and check for the output of the network. I already installed the NCS sdk on a Virtual Machine running ubuntu and went through the examples. It's all working. Now the problem is - how do I save a model I trained into the NCS? And then, how do I run some experiments with inputs to the network? Do you know some (very very simple) tutorials on it? I'm no computer scientist, so I would need a very simple step-by-step guide.
Thank you in advance!
Link Copied
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
@franciscoferreira34 We are currently working on a write up for what you're asking for. Other users have also asked for something similar in this thread: https://ncsforum.movidius.com/discussion/339/is-there-a-tutorial-or-documentation-for-running-custom-tensorflow-network-in-movidius#latest.
![](/skins/images/45C6C2D737ED71F4C51F0145C8CB1E9C/responsive_peak/images/icon_anonymous_message.png)
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page