Creating DataFrames from scratch

Usually, we create a DataFrame from an existing file or a database, but we can also create one from scratch. We can create a DataFrame from parallel lists of data.

How to do it...

1. Create parallel lists with your data in them. Each of these lists will be a column in the DataFrame, so they should have the same type:

   >>> import pandas as pd
   >>> import numpy as np
   >>> fname = ["Paul", "John", "Richard", "George"]
   >>> lname = ["McCartney", "Lennon", "Starkey", "Harrison"]
   >>> birth = [1942, 1940, 1940, 1943]
   

2. Create a dictionary from the lists, mapping the column name to the list:

   >>> people = {"first": fname, "last": lname, "birth": birth}
   

3. Create a DataFrame from the dictionary:

   >>> beatles = pd.DataFrame(people...