How to start Machine Learning?

Machine Learning now one of the trending technology in the 21st century and will take off high until the year 2020. By looking growth of machine learning it's a very hot topic for programmers. Every programmer wants to learn this new Technology, But it requires proper guidance since one can't start randomly and hope for best that eventually he'll become master in machine learning.
Here in this blog, I mentioned some basic step which is required for beginners i.e.who really  want to machine learning seriously:

The major topic which is required in machine learning which is the base of this technology:

1)Mathematics

   i)linear algebra
   ii)probability
   iii)statistics
   iv)calculas (Integration and Differential Equation for complex algorithm)
   v)graph
   vi)3-D space and vector 

2)Algorithm

  i)creating of the algorithm
  ii)Analysis of the algorithm

3)Programming

  i)Python (for beginners which I suggest)
  ii)R
  iii)MATLAB (for good visualization)

4)Imagination

  i)How you find the problem or imagine.
  ii)How you analyze or solve the problem(deciding which algorithm is best for this problem)
  iii)How you approach for a particular problem(how you gain high accuracy)

How to approach for all the above, as we know it is hard for those who hate mathematics, algorithm, programming or not good  at that particular subject but you still do good if you are good at imagination since as the technology evolves it becomes easier but  approach should  be right so here  I am explaining a few approaches which you  can take :

step-1:

 Machine Learning require less coding as these new frameworks and python library comes.
One should not just become stupid and start learning all the topics by-heart. If you know little knowledge of mathematics and can understand basic of all the topic mentioned above it's enough for beginners. As you go deep you can start studying that subject as much as your problem required since as we know we forgot as we don't touch a book for 2-3 days.

step-2:

Take any beginners course which is free and there is lots of website available where you can go and learn all the above-mentioned topic. I basically take one of the courses (i.e. udemy, udacity, datacamp, etc) since they teach full stack and you don't have to waste your time finding particular information in various website since this course provides good information.

step-3:

start doing small projects which can give you some confidence, try to apply the new algorithm on that projects and this will help you when you require more efficient models. You can take projects which are existing so if you go wrong you can refer the solution i.e. how they implement that particular problem.

step-4:

start coding problem statement and make an account on hackerrank, hackerearth, etc. this can build your logic developing strength and skills and making the new algorithm.

Thank You

Happy Machine Learning(coding)

Author:-Vikas Maurya