So you want to contribute to pandas by offering a patch? Excellent! This is the page you need to read. After you read the below guidelines, the first place to contribute issues & ideas to pandas is the GitHub Issue Tracker. You can filter using the “Community” label to see issues we believe are easy entry points for community contribution. Some longer discussions occur on the #pydata channel on, on the user or developer mailing list.

Contributing to the documentation

If you’re not the developer type, contributing to the documentation is still of huge value. You don’t even have to be an expert on pandas to do so! Something as simple as rewriting small passages for clarity as you reference the docs is a simple but effective way to contribute. The next person to read that passage will be in your debt!

Actually, there are sections of the docs that are worse off by being written by experts. If something in the docs doesn’t make sense to you, updating the relevant section after you figure it out is a simple way to ensure it will help the next person.

Once you have followed the steps below to download the source code from GitHub and set up your environment, make sure you have built the C extensions in place, then navigate to your local pandas/docs directory and run

python html

It will take awhile to build the first time, but subsequent builds only process the portions you’ve changed. Then just open the following file in a web browser:


And you’ll have the satisfaction of seeing your new and improved documentation!

The documentation is written in reStructuredText, which is almost like writing in plain English, and built using Sphinx. The Sphinx Documentation has an excellent introduction to reST. Review the Sphinx docs to perform more complex changes to the documentation as well.

Another way to help is to review docstrings. These can be edited directly in the codebase to make the functionality easier to understand.

Step-by-step overview

  1. Read carefully through the below guidelines on working with pandas code.
  2. Find a bug or feature you’d like to work on.
  3. Create a free account on GitHub, where we host our version controlled source repository.
  4. Set up your local development environment with git (Instructions).
  5. Fork the pandas repository (Instructions).
  6. Create a new working branch for your changes.
  7. Make sure your patch includes test coverage and performance benchmarks!
  8. Hook up Travis-CI
  9. Commit your changes and submit a pull request (Instructions).

Development Roadmap

  • (0.13) Improved SQL / relational database tools
  • Tools for working with data sets that do not fit into memory
  • (0.10) Better memory usage and performance when reading very large CSV files
  • Better statistical graphics using matplotlib
  • Integration with D3.js
  • Better support for integer NA values
  • Extend GroupBy functionality to regular ndarrays, record arrays
  • numpy.datetime64 integration, scikits.timeseries codebase integration. Substantially improved time series functionality.
  • ✔ Improved PyTables (HDF5) integration
  • NDFrame data structure for arbitrarily high-dimensional labeled data
  • ✔ Better support for NumPy dtype hierarchy without sacrificing usability
  • ✔ Add a Factor data type (in R parlance)

Code design and organization

File Hierarchy

  • pandas/core: Primary data structures (Series, DataFrame, ...) and related tools and algorithms
  • pandas/src: Cython and C code for implementing fundamental algorithms
  • pandas/io: Input/output tools (flat files, Excel, HDF5, SQL, ...)
  • pandas/tools: Auxiliary data algorithms: merge and join routines, concatenation, pivot tables, and more.
  • pandas/sparse: Sparse versions of Series, DataFrame, Panel
  • pandas/stats: Linear and panel regression, moving window regression. Will likely move to statsmodels eventually
  • pandas/util: Utilities, development, and testing tools
  • pandas/rpy: RPy2 interface for connecting to R


  • PEP8. We recommend using the flake8 tool for checking the style of your code.


Note that pandas is not 100% PEP8 compliant but we’re working on it. If you could help us toward this goal, it would be very helpful.

Working with the code

Version Control, Git, and GitHub

The code is hosted on GitHub. To contribute you will need to sign up for a free GitHub account. We use Git for version control to allow many people to work together on the project.

Some great resources for learning git:

Getting Started with Git

GitHub has instructions for installing git, setting up your SSH key, and configuring git.


You will need your own fork to work on the code. Go to the pandas project page and hit the fork button. You will want to clone your fork to your machine:

git clone pandas-yourname
cd pandas-yourname
git remote add upstream git://

This creates the directory pandas-yourname and connects your repository to the upstream (main project) pandas repository.

Creating a Branch

You want your master branch to reflect only production-ready code, so create a feature branch for making your changes. For example:

git branch shiny-new-feature
git checkout shiny-new-feature

This changes your working directory to the shiny-new-feature branch.

Making changes

Now hack away! Keep any changes in this branch specific to one bug or feature so it is clear what the branch brings to pandas.

Once you’ve made changes, you can see them by typing:

git status

If you’ve created a new file, it is not being tracked by git. Add it by typing

git add path/to/

Doing ‘git status’ again should give something like

# On branch shiny-new-feature
#       modified:   /relative/path/to/

Finally, commit your changes to your local repository with an explanatory message, such as

git commit -m "Optimized such-and-such function"

Your changes are now committed in your local repository.

Pushing your changes

When you want your changes to appear publicly on your GitHub page, push your forked feature branch’s commits

git push origin shiny-new-feature

Here origin is the default name given to your remote repository on GitHub. You can see the remote repositories

git remote -v

If you added the upstream repository as described above you will see something like

origin (fetch)
origin (push)
upstream        git:// (fetch)
upstream        git:// (push)

Now your code is on GitHub, but it is not yet a part of the pandas project. Before we get there, we need to address our testing and performance requirements for new code.


Test driven development

We’re serious about Test Driven Development (TDD). Any code you contribute must have adequate test coverage to be considered.

Like many packages, pandas uses the Nose testing system and the convenient extensions in numpy.testing.

Running the test suite

The best way to develop pandas is to build the C extensions in-place by running:

python build_ext --inplace

The tests can then be run directly inside your git clone (without having to install pandas) by typing:

nosetests pandas

Another very common option is to do a develop install of pandas:

python develop

This makes a symbolic link that tells the Python interpreter to import pandas from your development directory. Thus, you can always be using the development version on your system without being inside the clone directory.

How to write a test

The pandas.util.testing module has many special assert functions that make it easier to make statements about whether Series or DataFrame objects are equivalent. The easiest way to verify that your code is correct is to explicitly construct the result you expect, then compare the actual result to the expected correct result:

def test_pivot(self):
    data = {
        'index' : ['A', 'B', 'C', 'C', 'B', 'A'],
        'columns' : ['One', 'One', 'One', 'Two', 'Two', 'Two'],
        'values' : [1., 2., 3., 3., 2., 1.]

    frame = DataFrame(data)
    pivoted = frame.pivot(index='index', columns='columns', values='values')

    expected = DataFrame({
        'One' : {'A' : 1., 'B' : 2., 'C' : 3.},
        'Two' : {'A' : 1., 'B' : 2., 'C' : 3.}

    assert_frame_equal(pivoted, expected)

Performance testing with vbench

We created the vbench library to enable easy monitoring of the performance of critical pandas operations. These benchmarks are all found in the pandas/vb_suite directory. Interested users should simply look at the code there for the latest vbench API as vbench is still somewhat experimental and subject to change.

Contributing your changes to pandas

First, double check your code

When you’re ready to ask for a code review, you will file a pull request. Before you do, again make sure you’ve followed all the guidelines outlined in this document. You should also double check your branch changes against the branch it was based off of:

  1. Navigate to your repository on GitHub.
  2. Click on Branches.
  3. Click on the Compare button for your feature branch.
  4. Select the base and compare branches, if necessary. This will be master and shiny-new-feature, respectively.

Then, decide if you need to rebase

If you can avoid it, don’t rebase. But if there has been work in upstream/master related to the work in your branch since you started your patch, you may need to rebase.

A rebase replays commits from one branch on top of another branch to preserve a linear history. Remember, your commits may have been tested against an older version of master. If you rebase, you may introduce bugs. But if you don’t rebase, the two patches may conflict with each other!

Always make a new branch before doing rebase, and make sure you thoroughly understand rebasing lest you invoke the wrath of the git gods.

Finally, make the pull request

If everything looks good you are ready to make a pull request:

  1. Navigate to your repository on GitHub.
  2. Click on the Pull Request button.
  3. You can then click on Commits and Files Changed to make sure everything looks okay one last time.
  4. Write a description of your changes in the Preview Discussion tab.
  5. Click Send Pull Request.

This request then appears to the repository maintainers, and they will review the code. If you need to make more changes, you can make them in your branch, push them to GitHub, and the pull request will be automatically updated.

Optional: delete your merged branch

Once your feature branch is accepted into upstream, you’ll probably want to get rid of the branch. First, merge upstream master into your branch so git knows it is safe to delete your branch

git fetch upstream
git checkout master
git merge upstream/master

Then you can just do:

git branch -d shiny-new-feature

Make sure you use a lower-case -d, or else git won’t warn you if your feature branch has not actually been merged.

The branch will still exist on GitHub, so to delete it there do

git push origin --delete shiny-new-feature