Muhammad Raza
Hire me
python

My Favorite Pandas Tricks

Useful Pandas tricks for data analysis: finding unique values, filtering data with conditions, and saving results to CSV files.

Muhammad Raza · 2 min read #python

A few pandas tricks I reach for often when doing data analysis: finding unique values, filtering rows, grouping data, and saving results back to CSV.

  • Finding unique values
python
import pandas as pd
data = pd.read_csv('https://gist.githubusercontent.com/tiangechen/b68782efa49a16edaf07dc2cdaa855ea/raw/0c794a9717f18b094eabab2cd6a6b9a226903577/movies.csv')
data.Film.unique()

This will print only unique values in Film column in the csv.

  • Filtering Data

Say you only want movies with an audience score above 50 that are comedies. Filtering handles this in one line.

python
new_data = data[(data['Audience score %'] > 50) & (data.Genre == 'Comedy')]
  • Saving to csv.

Pandas has a function that saves data to a csv file. For instance, to save all the unique movie names, first convert them to a data frame:

python
uniq = data.Film.unique()
out = pd.DataFrame(uniq)
out.to_csv('uniq.csv')

This will create a csv file of unique names.

  • Groupby

Groups data by a column. For instance, to count movies by genre, use groupby.

python
data.groupby('Genre').Film.agg(['count'])

This outputs the total number of movies for each genre. You can also use other aggregations like sum, mean, and median.

  • String Operations

You can also use string operations when working with text data.

python
# lower case a specific column

data['Genre'] = data['Genre'].str.lower()

This lowercases the Genre column. You can also use upper() for uppercase, or apply your own regex with replace.

Let me know in the comments what your favorite pandas trick is.