Getting started

Installation

Working with conda?

pandas is part of the Anaconda distribution and can be installed with Anaconda or Miniconda:

Prefer pip?

pandas can be installed via pip from PyPI.

In-depth instructions?

Installing a specific version? Installing from source? Check the advanced installation page.

Intro to pandas

When working with tabular data, such as data stored in spreadsheets or databases, pandas is the right tool for you. pandas will help you to explore, clean and process your data. In pandas, a data table is called a DataFrame.

../_images/01_table_dataframe.svg

pandas supports the integration with many file formats or data sources out of the box (csv, excel, sql, json, parquet,…). Importing data from each of these data sources is provided by function with the prefix read_*. Similarly, the to_* methods are used to store data.

../_images/02_io_readwrite.svg

Selecting or filtering specific rows and/or columns? Filtering the data on a condition? Methods for slicing, selecting, and extracting the data you need are available in pandas.

../_images/03_subset_columns_rows.svg

pandas provides plotting your data out of the box, using the power of Matplotlib. You can pick the plot type (scatter, bar, boxplot,…) corresponding to your data.

../_images/04_plot_overview.svg

There is no need to loop over all rows of your data table to do calculations. Data manipulations on a column work elementwise. Adding a column to a DataFrame based on existing data in other columns is straightforward.

../_images/05_newcolumn_2.svg

Basic statistics (mean, median, min, max, counts…) are easily calculable. These or custom aggregations can be applied on the entire data set, a sliding window of the data or grouped by categories. The latter is also known as the split-apply-combine approach.

../_images/06_groupby.svg

Change the structure of your data table in multiple ways. You can melt() your data table from wide to long/tidy form or pivot() from long to wide format. With aggregations built-in, a pivot table is created with a single command.

../_images/07_melt.svg

Multiple tables can be concatenated both column wise as row wise and database-like join/merge operations are provided to combine multiple tables of data.

../_images/08_concat_row.svg

pandas has great support for time series and has an extensive set of tools for working with dates, times, and time-indexed data.

Data sets do not only contain numerical data. pandas provides a wide range of functions to clean textual data and extract useful information from it.

Coming from…

Are you familiar with other software for manipulating tablular data? Learn the pandas-equivalent operations compared to software you already know:

R project logo

The R programming language provides the dataframe data structure and multiple packages, such as tidyverse use and extend data.frames for convenient data handling functionalities similar to pandas.

SQL logo

Already familiar to SELECT, GROUP BY, JOIN, etc.? Most of these SQL manipulations do have equivalents in pandas.

STATA logo

The data set included in the STATA statistical software suite corresponds to the pandas dataframe. Many of the operations known from STATA have an equivalent in pandas.

SAS logo

The SAS statistical software suite also provides the data set corresponding to the pandas dataframe. Also SAS vectorized operations, filtering, string processing operations, and more have similar functions in pandas.

Tutorials

For a quick overview of pandas functionality, see 10 Minutes to pandas.

You can also reference the pandas cheat sheet for a succinct guide for manipulating data with pandas.

The community produces a wide variety of tutorials available online. Some of the material is enlisted in the community contributed Community tutorials.