I assume you are referring to the beginners course to machine learning and not his deep learning specialisation. To be fair a lot of the courses on Udemy are not worth the money as they are of poor quality. I took the Andrew Ng ML course a few years ago. At the time the course practicals were to be done using the free software Octave. Now as good as Octave is it is not an industry used software. So upon completing the course I would probably spend a little time doing all the practical exercises in Python. So learn a little Scikit-learn and Numpy in the process. Download the package Anaconda Python/R Distribution – Free Download
then install and get to work doing the exercises again in Python. This way it will consolidate all the basics and also will familiarise yourself with some of the popular tools used in the machine learning field. I would highly recommend this book as it does exactly that. It develops from first principles all the algorithms that Andrew Ng discusses in his course.
Machine Learning Books – Machine Learning Nomad
If you don’t feel confident enough to do this at this stage due to poor or no Python skills then I would suggest initially you go to the Welcome to Python.org website and work your way through some of the tutorials in combination with working on small Python coding projects to build skills on coding and testing. I for example wrote a chess program in object oriented Python to teach myself the more intricate features of Python. It was fun and challenging. Another great site for improving coding skills is Codewars: Train your coding skills. I love this site as it presents a fun way to challenge yourself on all areas of coding and not just in Python too if you prefer other languages. Once you are comfortable enough with Python then move back to the machine learning book mentioned above.
The above are free resources. However if you think a paid certificated course is more valuable then I can recommend Datacamp
Learn R, Python & Data Science Online
as I have an active subscription with them I feel I can talk about them with confidence. Datacamp is a great way to build up the initial foundations of what you need to know for a career in data science. They present course information as video instruction on a topic followed by guided practice in an integrated programming environment where you are only expected to fill in code stubs and not have to worry about setting up the environment. Which as you will learn is challenging when you first start especially if you have little experience of programming. This “hand holding” allows you to focus on the machine learning and not have to go through the hair pulling agony of fiddling with a real coding platform such as Netbeans or Anaconda etc. Datacamp have several skills tracks you can follow i.e. for R and for Python. There are hundreds of courses included in the subscription as well as larger practical projects which can be done in order to take on more challenging coding. See my other post on which courses are available currently
Machine Learning Nomad’s answer to E-learning: What are some good DataCamp courses?
Leave a Reply