In this course, we are going to focus on Exploratory Data Analysis techniques for machine learning projects.
What is EDA?
Exploratory Data Analysis (EDA) is the first approach to a dataset before applying any transformation or model. It's able to make us analyze a dataset just as it is, graphically exploring the correlation between the features and their predictive power. In fact, EDA makes use of data visualization concepts to extract the information hidden inside data.
Why is EDA useful?
EDA helps us explore the information hidden inside a dataset before applying any model or algorithm. So, it's bias-free. Moreover, it lets us figure out whether our features have a predictive power or not, so if the machine learning project we are working on has chances to be successful. Without EDA, we may give the wrong data to a model without reaching any success.
Is EDA a good skill to have?
Definitely yes. The results of a good EDA can be a deliverable by themselves and can let managers understand information before spending time and money on models and algorithms. EDA makes data scientists do their original job: extract business information from data.
What will I learn with this course?
Completing this course you will learn the basic principles of Exploratory Data Analysis, including pair plots, histograms, conditional histograms, some powerful Python libraries and other visualization tools. Every lesson is made by a practical example in Python programming language using Jupyter notebooks.