Today I wanted implement a deep learning model in openvino. So I went to the page here and started reading. After starting the reading, I struggle to understand things and started to deviate a lot from my objective. For example:

Speeds time-to-market via an easy-to-use library of computer vision functions and pre-optimized kernels

what are pre-optimized kernels?

Includes optimized calls for computer vision standards including OpenCV*, OpenCL™, and OpenVX*. What are optimised calls?

Intel® Media SDK Offers access to hardware accelerated video codecs and frame processing

what is SDK?

A set of simple console applications demonstrating how to use the Inference Engine in your applications

what is console appplications,inference engine?

I am not a CS graduate. What can I do to improve, should I take a CS curriculum? Sometimes I exactly don't understand certain things, but I proceed with a rough idea. Am I trying to understand in detail here? I don't know, because I am perfectionist. Else what can I do to improve?

EDIT: The related answer is quite good. But I wanted to know, how to approach a concept where we need to understand the theory. Sometimes you may need to understand what is happening behind it before implementing it. What can one do in such cases?

  • 1
    $\begingroup$ We rarely close questions as duplicate on this site, but I admit that the two questions seem almost entirely identical. We're open to re-opening the question, but could you add some additional information that makes clearer how this question is different from the one Flater linked to? $\endgroup$ – Ben I. Jan 22 '20 at 14:14

Here are two pianists

Do you realize no 2 was once no 1?

Here's an In between stage

To get a handle on the issue you need ten thousand hour rule

You can reduce 10,000 hours to 1000 hours and go from world class to normal professional. Perhaps 1000 to 100 and go from professional to modestly proficient.

But even@100 that's about classroom time of two typical university courses.

Put another way musical proficiency takes about 10-15 years of work, medical 5-10. Computer/software takes 3-4. Maybe you can reduce 3 to 1.

Have you put in that 1 gruelling year of work?

  • $\begingroup$ A bit esoteric but true. However you could improve the answer by saying more (and linking less). It is completely understandable to those who already know this, but not so much to novices. In Tai Chi we say that it takes 10,000 repetitions of a move to know it. $\endgroup$ – Buffy Jan 22 '20 at 12:18
  • $\begingroup$ The 10,000 hour rule is to become world-class in expertise or proficiency, not to "get a handle" on something. $\endgroup$ – Erik Eidt Jan 22 '20 at 15:57
  • $\begingroup$ @EricEidt I think you've misunderstood my use of ‘handle’. When I said "get a handle" I was not meaning it at the (object) level of "How do I get a handle on ML?" but the meta level of "I dont understand what's holding me back..." Which is after all the central point of the q (AIUI) $\endgroup$ – Rusi Jan 22 '20 at 16:40

Not the answer you're looking for? Browse other questions tagged or ask your own question.