swifter_parallel_apply_function_to_pandas_dataframe.py

1import pandas as pd
2import swifter
3
4# Define a sample function to apply
5def filter_function(x):
6    return x**2
7
8# Create a sample dataframe
9df = pd.DataFrame({'a': [1, 2, 3, 4, 5] * 100000})
10
11# Use swifter to apply the function to a column
12# Swifter automatically decides whether to use Dask, Ray, or vectorized Pandas for speed
13df['b'] = df['a'].swifter.apply(filter_function)
14
15# Use swifter to apply the function to a series
16series = pd.Series([1, 2, 3, 4, 5] * 100000)
17result = series.swifter.apply(filter_function)
18
19print(df.head())