How to use `pipe` in pandas

This is the code before.

import pandas as pd

df = pd.read_csv('https://calmcode.io/datasets/bigmac.csv')

df2 = (df
  .assign(date=lambda d: pd.to_datetime(d['date']))
  .sort_values(['currency_code', 'date'])
  .groupby('currency_code')
  .agg(n=('date', 'count')))

df.loc[lambda d: d['currency_code'].isin(df2[df2['n'] >= 32].index)]

This is the code after.

import pandas as pd

df = pd.read_csv('https://calmcode.io/datasets/bigmac.csv')

def set_dtypes(dataf):
    return (dataf
            .assign(date=lambda d: pd.to_datetime(d['date']))
            .sort_values(['currency_code', 'date']))

def remove_outliers(dataf):
    min_row_country=32
    countries = (dataf
                .groupby('currency_code')
                .agg(n=('name', 'count'))
                .loc[lambda d: d['n'] >= min_row_country]
                .index)
    return (dataf
            .loc[lambda d: d['currency_code'].isin(countries)])

df.pipe(set_dtypes).pipe(remove_outliers)

Notice how we now have a pipeline with steps. Each function is like a lego brick, which introduces a lot of structure to our code!