I would love to be part of it, where/how to start?

Hello All,

I recently discovered “almost by accident” the product Jetson TX1. All the reading about the product, videos lead me to also read about the deep learning and neural networks, I have to say, I’m fascinated.
Knowing that my knowledge as a developer=0, I have some IT background, and I would love to be able to “Play “with the Jetson TX1, for the robotic (I watched some videos from Jetsonhacks YouTube channel)
I’d like to start with a Jetson Tx1 and a Kinect, to understand and train myself

What is your background?
What are first steps?
Which language(s) I need to learn?
Where/how to learn about deep learning and neural networks from the very beginning?

Please be patient with me and thank you very much for any help/information/suggestion in this new adventure for me

Great question! I would start by browsing the available documentation (like the Jetson TX1 Datasheet and Carrier Spec) from here: https://developer.nvidia.com/embedded/downloads#?tx=$product,jetson_tx1

See here for our OpenCV and VisionWorks tutorials: https://developer.nvidia.com/embedded/learn/tutorials
We will be posting more episodes of the OpenCV tutorial in the next few days. The VisionWorks tutorials are also good to learn, because the performance is optimized for Jetson TX1.

NVIDIA has online courses available in Deep Learning, including how to get started using DL frameworks like Caffe. https://developer.nvidia.com/deep-learning-courses

There’s a lot out there to learn about deep learning and networks, here are a few good sites which contain links to more reading material:


I also often use Google Scholar (http://scholar.google.com) when researching DL.

To get an idea of what people are using Jetson TX1 for so far, check out our Success Story videos, posted here: https://developer.nvidia.com/embedded/learn/success-stories

If you are interested in aquiring a Jetson TX1 Developer Kit, you can purchase it from here (https://developer.nvidia.com/embedded/buy/jetson-tx1-devkit) and apply for EDU discount (if you have .EDU email address). The devkit is on backorder right now but NV website is replenishing units shortly for fulfillment.