Data exploration
Data exploration is generally performed by presenting a meaningful synthesis of its distribution—it could be through a sequence of graphs, by describing it with a set of numerical parameters, or by approximating it with simple functions. Now let's explore different possibilities, and how to accomplish them with different tools in the SciPy stack.
Picturing distributions with graphs
The type of graph depends on the type of variable (categorical, quantitative, or dates).
Bar plots and pie charts
When our data is described in terms of categorical variables, we often use pie charts or bar graphs to represent it. For example, we access the Consumer Complaint Database from the Consumer Financial Protection Bureau, at http://catalog.data.gov/dataset/consumer-complaint-database ...
Get Mastering SciPy now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
