The progress that machine learning has made in past decade strikes everyone as genuine and astounding. Tons of libraries, architectures and mathematical equations have been developed to support machine learning. Such a growth is attracting every developer. No matter even if he has been doing quantum computing, he might want to learn machine learning. It might generate some insightful prediction for him. But the most commonest question I face when I meet new developers is to do with getting started.

Being a developer you might be having good programming skills, or may be not. Programming skills can be evolved with time and practice. 1 question a day will do the job. But before jumping into machine learning one needs to make sure you are good at maths -Algebra, Vectors, Matrices, derivatives specifically. You cannot learn maths in a day. So here is the tip.

Tip 1:Pick up a problem from the Engineering Mathematics course of past, and try to dive into in it with practical examples. Then try to solve these equations using scientific computing libraries such as scipy and numpy.

If you are comfortable with the above stuff, you might want to get a headstart to ML, so you plan to take the MOOCs on coursera. Machine Learning courses by Dr. Andrew Ng are most popular these days. But then you might fail to complete those courses due to lack of time to complete the assignments. Here is the tip for you.

Tip 2:You can just view the videos rather completing the assignments along at first. Once you are comfortable with the theory behind machine learning, you can start doing the assignments. Python Implementations of the algorithms are the best .

While doing freelancing, I frequently encounter people who are good with first two points, but they struggle at writing the python implementations. They just lack the python packages knowledge and practice over them. Here are some resource that will help you:

Tip 3:Pandasâ€Šâ€”â€ŠA Data Processing Library In PythonUnderstanding the essence of Linear AlgebraNumpyâ€Šâ€”â€ŠA Numerical Computing Library In PythonUnderstanding Neural Networks DeepLearningBest DeepLearning MOOC on Coursera

Now, you have worked hard over yourself in learning the concepts of deep learning, you want to build your own projects and a portfolio to get a good job. Here is a list of simple projects I worked on currently.

Tip 4:Dogs Breed Classification (uses transfer Learning)Optical Character Recognition (uses sequence models)See and Tell Using Deep LearningAsk the Net Face Swapping using DeepLearningVoice Cloning Using DeepLearning

At last, the important tip, that will really put your programming, mathematical and data science skills to the test

Tip 5:Do not just git clone the projects, but also implement your own architectures and hacks from scratch. Machine Learning is all about the research, so doing a good research will boost up your confidence.

**BEST OF LUCK !!!**