The availability of affordable compute power enabled by Moore’s law has been enabling rapid advances
in Machine Learning solutions and driving adoption across diverse segments of the industry. The ability
to learn complex models underlying the real-world processes from observed (training) data through
systemic, easy-to-apply Machine Learning solution stacks has been of tremendous attraction to businesses
to harness meaningful business value. The appeal and opportunities of Machine Learning have resulted in
the availability of many resources—books, tutorials, online training, and courses for solution developers,
analysts, engineers, and scientists to learn the algorithms and implement platforms and methodologies. It
is not uncommon for someone just starting out to get overwhelmed by the abundance of the material. In
addition, not following a structured workflow might not yield consistent and relevant results with Machine
Learning solutions.
1