-
Notifications
You must be signed in to change notification settings - Fork 0
Google Summer of Code
This is the GSoC'16 ideas page for pandas.
pandas is a Python package providing fast, flexible, and expressive data structures designed to make working with “relational” or “labeled” data both easy and intuitive. It aims to be the fundamental high-level building block for doing practical, real world data analysis in Python. Additionally, it has the broader goal of becoming the most powerful and flexible open source data analysis / manipulation tool available in any language. It has become a centerpiece of the PyData stack.
This page lists a number of ideas for Google Summer of Code projects for pandas, plus gives some pointers for potential GSoC students on how to get started with contributing and putting together their application.
Pandas participates in GSoC 2016 under the umbrella of Python Software Foundation / NUMFocus.
PSF student guidelines: http://wiki.python.org/moin/SummerOfCode/Expectations
Advice on writing a proposal (written with the Mailman project in mind, but generally applicable): http://wiki.list.org/DEV/SPAM
We expect from students that they're at least comfortable with Python (intermediate level). Some projects may also require Cython or C/C++ skills. Knowing how to use Git is also important; this can be learned before the official start of GSoC if needed though.
Potential candidates should to take a look at the guidelines on how to contribute to pandas, see the documentations here. Making a small enhancement/bugfix/documentation fix/etc (does not need to be related to your proposal) to pandas before applying for the GSoC is a requirement from the PSF; it can help you get some idea how things would work during the GSoC.
Start on your proposal early, post a draft to the mailing list and iterate based on the feedback you receive. This will not only improve the quality of your proposal, but also help you find a suitable mentor.
| Difficulty | Involves |
|---|---|
| Intermediate | python, cython, c, c++ |
Add to_r_data_set, from_r_data_set or similar, a performant interface to reading/writing R datasets directly (to/from disk). Some references here, and here, and here, and conversations here and here. Couple of issues to deal with. picklr is not license compatible, though uses xdr which would be nice as its a built in python module. Whereas ReadStat would need substantial wrapping. Might be easy to translate the julia code almost directly to cython. Mentor will be devasena prasad
|
| Difficulty | Involves |
|---|---|
| Intermediate | python, cython |
| Implement some missing features in sparse, see the open issues here. Sparse needs some TLC in interactions with other pandas objects and compatibility. |
| Difficulty | Involves |
|---|---|
| Novice | python, cython |
see issue here, basically adding a weights arg for things like .mean, etc. This is a relatively straightforward project and would be suitable for a newcomer to pandas. |
| Difficulty | Involves |
|---|---|
| Intermediate | python, cython, cyavor, parquet-cpp |
Adding input-output connectors & support for to_* and from_* for these binary formats (to use existing libraries to actually read/write; this item is for integration/shipping within pandas). Requires some knowledge of the outside library and a bit of pandas internals. These libraries will allow pandas to interact better with some of the big data eco-system. These could be done as multiple independent projects. |
- avro support
- parquet support
- BSON?
- RData
| Difficulty | Involves |
|---|---|
| Intermediate | python, cython, numba |
- Construction of a general interface (possibly via a
numbaextension), to allow automatic direction of code tonumbavia.applys. - Ahead-of-time to generate code for groupby (and other algos), rather than direct templating. These are two separate projects and requires some pandas internal knowledge as well as familiarity with
numba
| Difficulty | Involves |
|---|---|
| Advanced | python, cython |
- Implement as sub-class of
IntervalIndex - Make an extension dtype, and integrate as a first class object in the pandas. A pretty deep knowledge of pandas internals is needed. This could actually be two separate somewhat independent projects.
| Difficulty | Involves |
|---|---|
| Advanced | python, cython |
Support for non-ns dtypes. This requires deep knowledge of pandas internals is needed. It would be helpful to understand numpy dtype mechanisms. A master issue is here
|
| Difficulty | Involves |
|---|---|
| Advanced | python, cython |
Construct a new dtype to support string operations (moved from object). An understanding of Categorical is required, as well as deep knowledge of pandas internals is needed. It would be helpful to understand numpy dtype mechanisms. |
| Difficulty | Involves |
|---|---|
| Advanced | python |
| Allow more meta-data to be attached to pandas objects, and propogate in the common cases. here is a discussion of this |
| Difficulty | Involves |
|---|---|
| Intermediate | python, pandas, xarray |
Flesh out additional features that are needed to fully support Panel operations and implement in xarray. See if can port some selected operations from xarray back to pandas. This would be a relatively straightforward project and would be suitable for a newcomer to pandas. |
| Difficulty | Involves |
|---|---|
| Intermediate | python, pandas, matplotlib |
Deep knowledge of pandas internals is needed. It would be helpful to understand numpy dtype mechanisms. |
|
There would be possible collaboration with matplotlib. here is more of a discussion
|
| Difficulty | Involves |
|---|---|
| Novice | python, pandas |
| a list of novice level bug fixes / enhancements |
| Difficulty | Involves |
|---|---|
| Novice | python, pandas |
| a list of intermediate level bug fixes / enhancements |
| Difficulty | Involves |
|---|---|
| Advanced | python, pandas, cython |
| a list of advanced level bug fixes / enhancements |
| Difficulty | Involves |
|---|---|
| Advanced | python, pandas, c++ |
| An internal refactor of pandas preserving the user-facing API and the developer API (numpy). requires a deep knowledge of c++ and pandas internals. |