# Studying Machine Learning

A student wrote to me: “I want to learn more about machine learning and have looked at several tutorial websites but I am confused and do not know how to start. Please advise.”

Answer: “Before learning something new, you need to have a plan to go from the basics to the advance. If you start with difficult subjects or tools, you may be confused. I know some students get discouraged and quit because they are jumping into something beyond their capability.

Before learning machine learning you need to have good programming skill. I think Python is probably one of the best languages to start. Besides writing code in Python, you also need to learn about Python Libraries since there are many of them. You need to be familiar with Numpy, the fundamental package for scientific computing in Python. You need to know Pandas to collect, organize and prepare the data for your algorithms and Matplotlib to plot Mathematically operations Visually in Dimensions. You can learn these from Udemy:

www.udemy.com/deep-learning-prerequisites-the-numpy-stack-in-python/?siteID=JVFxdTr9V80-Jew0L0NIvkOlKDwe2NWy8g&LSNPUBID=JVFxdTr9V80

After learning Python and familiar with Python libraries and tools, you need to review your math skills because machine learning is based on the knowledge of maths such as Algebra, Calculus, Probability, and Statistics.

Having the good foundation (i.e., Python and math skills) I recommend that you take a MOOCs course (i.e. Coursera) taught by Andrew Ng. to understand all the theories of Machine learning. This is the best course about machine learning that I have seen: https://www.coursera.org/learn/machine-learning

By completing the above recommendations, I think you have a strong foundation to learn any machine learning courses or tools to start your career.

## Sources

- Blogs of Prof. John Vu, Carnegie Mellon University