Do not decelerate your data science journey.

Original article can be found here (source): Artificial Intelligence on Medium

Do not decelerate your data science journey.

Data Scientist: Often coined as ‘sexiest job of the 21st century’ is the demand of the hour. Any big firm would now need a data scientist, data analyst for handling and extracting out data.

In order to outrun your competitors, you immediately need to realise the following points:

1. Spending too much time on theory.

The beginners get into the trap of spending too much time on theory, be it maths or machine learning. This approach is inefficient because you won’t retain the concepts, Data science is an applied field and the best way to solidify skills is by practicing. There’s a greater risk that you’ll become demotivated and give up if you don’t see how the subject connects to the real world.

To avoid this mistake:

· Balance your studies with projects that provide you hands-on practice.

· Learn to be comfortable with practical knowledge.

· Learn how each piece fits into the big picture.

2. Coding too many algorithms from scratch.

This mistake causes students to miss the forest for the trees. In the beginning, you really don’t need to code every algorithm from scratch. While it’s good to implement a few for learning, the reality states that algorithms are becoming commodities. Credits to advance machine learning libraries and cloud-based solutions, most practitioners actually never code algorithms from scratch. It’s more important to implement the right algorithms in the right settings.

To avoid this mistake:

· Pick up general-purpose machine learning libraries like Scikit-Learn .

· If you do code an algorithm from scratch, do so with the intention of learning instead of perfecting your implementation.

· Understand the landscape of modern machine learning algorithms and their strengths and weaknesses.

3. Jumping into the deep end.

People enter this field with the intention building the technology of the future: Self-Driving Cars, Advanced Robotics, Computer Vision, and so on. These are achieved by techniques such deep neural networks and natural language processing. However, it’s important to master the fundamentals.

To avoid this mistake:

· Master the techniques and algorithms of “classical” machine learning, which serve as building blocks for advanced topics.

· Know that classical machine learning still has incredible untapped potential. Though the algorithms are already mature, we are still in the earliest stage of discovering fruitful ways to use them.

· Learn a systematic approach to structuring machine learning project.