Analyze some basic car data
Level: Beginner (score: 2)
In this exercise you will analyze some basic car data. Here is the (fake) JSON data we created with Mockeroo - snippet below / full output here:
[{"id":1,"automaker":"Dodge","model":"Ram Van 1500","year":1999}, {"id":2,"automaker":"Chrysler","model":"Town & Country","year":2002}, {"id":3,"automaker":"Porsche","model":"Cayenne","year":2008}, ... 997 car entries more ... ]
First you will write most_prolific_automaker
to find out which automaker produces the most new models for a particular year.
Secondly you will write get_models
which filters the data set down to car models produced by a particular automaker and year (as passed into the function).
To keep it a Beginner Bite we'll pause here, but if you like this data set, let us know and we make a follow-up Bite, maybe we can add some financial data :)
Check out the docstrings and pytests and give it a shot. Good luck and keep calm keep and code in Python.