Check CSV Equivalence Using Polars

Today I Learned added on 2025-11-20, learned on 2025-11-20; edited on 2026-06-14.

I found myself reviewing two PRs made by the same individual and each PR had a csv file containing forecasts. The csvs seemed identical, but I wasn’t sure, so I wanted a way to check for equivalence using Python.

Given that, at present, I prefer polars over pandas, I searched polars for a method and found the method equals (see https://docs.pola.rs/api/python/stable/reference/dataframe/api/polars.DataFrame.equals.html) in the DataFrame class .

Example csv file as example_01.csv:

column_1,column_2
45,98
32,31
17,100

Example csv file as example_02.csv:

column_1,column_2
87,98
32,31
17,100

First, the non-Python, Unix command diff (the standard method I’ve employed for equivalence testing) can also be used.

diff example_01.csv example_02.csv

# output:
# 2c2
# < 45,98
# ---
# > 87,98

In Python, using the equals method of polars, which outputs True or False, instead of the actual differences if there are any:

import polars as pl


df_01 = pl.read_csv("example_01.csv")
df_02 = pl.read_csv("example_02.csv")

# output: False
df_01.equals(df_02)