Turn a matrix of data to a polars dataframe

from_records

In data science and analysis, storing data in an array or nested arrays (a matrix) is a common practice. Unfortunately, performing operations directly on arrays can be challenging. Fortunately, Polars allows you to convert a matrix into a dataframe. Below is a matrix of people and their professions.

data_matrix = [['Joram','Ashwin','Ollie','Jeremie'], ['Data Consultant','Product Manager','Lawyer','Doctor']]
data_matrix

[['Joram', 'Ashwin', 'Ollie', 'Jeremie'],
 ['Data Consultant', 'Product Manager', 'Lawyer', 'Doctor']]

Turn matrix to dataframe

Polars provides the from_records expression, which creates a dataframe from matrix records. You can also add a schema to specify column names for the dataframe. Here’s how to use it:

import polars as pl
df = pl.from_records(data_matrix, schema=['Name', 'Profession'])
df

shape: (4, 2)

Name	Profession
str	str
"Joram"	"Data Consultant"
"Ashwin"	"Product Manager"
"Ollie"	"Lawyer"
"Jeremie"	"Doctor"

Join my Polars course to improve your data analysis skills.