Welcome to the DataFrames.jl documentation!
This resource aims to teach you everything you need to know to get up and running with tabular data manipulation using the DataFrames.jl package.
For more illustrations of DataFrames.jl usage, in particular in conjunction with other packages you can check-out the following resources (they are kept up to date with the released version of DataFrames.jl):
- DataFrames.jl: Flexible and Fast Tabular Data in Julia article published in the Journal of Statistical Software
- Data Wrangling with DataFrames.jl Cheat Sheet
- DataFrames Tutorial using Jupyter Notebooks
- Julia Academy DataFrames.jl tutorial
- JuliaCon 2019, JuliaCon 2020, JuliaCon 2021, JuliaCon 2022, PyData Global 2020, and ODSC Europe 2021 tutorials
- DataFrames.jl showcase
If you prefer to learn DataFrames.jl from a book you can consider reading:
DataFrames.jl provides a set of tools for working with tabular data in Julia. Its design and functionality are similar to those of pandas (in Python) and
data.table and dplyr (in R), making it a great general purpose data science tool.
DataFrames.jl plays a central role in the Julia Data ecosystem, and has tight integrations with a range of different libraries. DataFrames.jl isn't the only tool for working with tabular data in Julia – as noted below, there are some other great libraries for certain use-cases – but it provides great data wrangling functionality through a familiar interface.
To understand the toolchain in more detail, have a look at the tutorials in this manual. New users can start with the First Steps with DataFrames.jl section.
You may find the DataFramesMeta.jl package or one of the other convenience packages discussed in the Data manipulation frameworks section of this manual helpful when writing more advanced data transformations, especially if you do not have a significant programming experience. These packages provide convenience syntax similar to dplyr in R.
If you use metadata when working with DataFrames.jl you might find the TableMetadataTools.jl package useful. This package defines several convenience functions for performing typical metadata operations.
The Julia data ecosystem can be a difficult space for new users to navigate, in part because the Julia ecosystem tends to distribute functionality across different libraries more than some other languages. Because many people coming to DataFrames.jl are just starting to explore the Julia data ecosystem, below is a list of well-supported libraries that provide different data science tools, along with a few notes about what makes each library special, and how well integrated they are with DataFrames.jl.
- StatsKit.jl: A convenience meta-package which loads a set of essential packages for statistics, including those mentioned below in this section and DataFrames.jl itself.
- Statistics: The Julia standard library comes with a wide range of statistics functionality, but to gain access to these functions you must call
- LinearAlgebra: Like
Statistics, many linear algebra features (factorizations, inversions, etc.) live in a library you have to load to use.
- SparseArrays are also in the standard library but must be loaded to be used.
- FreqTables.jl: Create frequency tables / cross-tabulations. Tightly integrated with DataFrames.jl.
- HypothesisTests.jl: A range of hypothesis testing tools.
- GLM.jl: Tools for estimating linear and generalized linear models. Tightly integrated with DataFrames.jl.
- StatsModels.jl: For converting heterogeneous
DataFrameinto homogeneous matrices for use with linear algebra libraries or machine learning applications that don't directly support
DataFrames. Will do things like convert categorical variables into indicators/one-hot-encodings, create interaction terms, etc.
- MultivariateStats.jl: linear regression, ridge regression, PCA, component analyses tools. Not well integrated with DataFrames.jl, but easily used in combination with
- Machine Learning
- MLJ.jl: if you're more of an applied user, there is a single package the pulls from all these different libraries and provides a single, scikit-learn inspired API: MLJ.jl. MLJ.jl provides a common interface for a wide range of machine learning algorithms.
- ScikitLearn.jl: A Julia wrapper around the full Python scikit-learn machine learning library. Not well integrated with DataFrames.jl, but can be combined using StatsModels.jl.
- AutoMLPipeline: A package that makes it trivial to create complex ML pipeline structures using simple expressions. It leverages on the built-in macro programming features of Julia to symbolically process, manipulate pipeline expressions, and makes it easy to discover optimal structures for machine learning regression and classification.
- Deep learning: KNet.jl and Flux.jl.
- Plots.jl: Powerful, modern plotting library with a syntax akin to that of matplotlib (in Python) or
plot(in R). StatsPlots.jl provides Plots.jl with recipes for many standard statistical plots.
- Gadfly.jl: High-level plotting library with a "grammar of graphics" syntax akin to that of ggplot (in R).
- AlgebraOfGraphics.jl: A "grammar of graphics" library build upon Makie.jl.
- VegaLite.jl: High-level plotting library that uses a different "grammar of graphics" syntax and has an emphasis on interactive graphics.
- Plots.jl: Powerful, modern plotting library with a syntax akin to that of matplotlib (in Python) or
- Data Wrangling:
- Impute.jl: various methods for handling missing data in vectors, matrices and tables.
- DataFramesMeta.jl: A range of convenience functions for DataFrames.jl that augment
transformto provide a user experience similar to that provided by dplyr in R.
- DataFrameMacros.jl: Provides macro versions of the common DataFrames.jl functions similar to DataFramesMeta.jl, with convenient syntax for the manipulation of multiple columns at once.
- Query.jl: Query.jl provides a single framework for data wrangling that works with a range of libraries, including DataFrames.jl, other tabular data libraries (more on those below), and even non-tabular data. Provides many convenience functions analogous to those in dplyr in R or LINQ.
- You can find more information on these packages in the Data manipulation frameworks section of this manual.
- And More!
- DataFrames.jl work well with a range of formats, including:
While not all of these libraries are tightly integrated with DataFrames.jl, because
DataFrames are essentially collections of aligned Julia vectors, so it is easy to (a) pull out a vector for use with a non-DataFrames-integrated library, or (b) convert your table into a homogeneously-typed matrix using the
Matrix constructor or StatsModels.jl.
DataFrames.jl is a great general purpose tool for data manipulation and wrangling, but it's not ideal for all applications. For users with more specialized needs, consider using:
- TypedTables.jl: Type-stable heterogeneous tables. Useful for improved performance when the structure of your table is relatively stable and does not feature thousands of columns.
- JuliaDB.jl: For users working with data that is too large to fit in memory, we suggest JuliaDB.jl, which offers better performance for large datasets, and can handle out-of-core data manipulations (Python users can think of JuliaDB.jl as the Julia version of dask).
Note that most tabular data libraries in the Julia ecosystem (including DataFrames.jl) support a common interface (defined in the Tables.jl package). As a result, some libraries are capable or working with a range of tabular data structures, making it easy to move between tabular libraries as your needs change. A user of Query.jl, for example, can use the same code to manipulate data in a
Table (defined by TypedTables.jl), or a JuliaDB table.
If there is something you expect DataFrames to be capable of, but cannot figure out how to do, please reach out with questions in Domains/Data on Discourse. Additionally you might want to listen to an introduction to DataFrames.jl on JuliaAcademy.
Please report bugs by opening an issue.
You can follow the source links throughout the documentation to jump right to the source files on GitHub to make pull requests for improving the documentation and function capabilities.
Please review DataFrames contributing guidelines before submitting your first PR!
Information on specific versions can be found on the Release page.
- First Steps with DataFrames.jl
- Getting Started
- Database-Style Joins
- The Split-Apply-Combine Strategy
- Reshaping and Pivoting Data
- Categorical Data
- Missing Data
- Data manipulation frameworks
Only exported (i.e. available for use without
DataFrames. qualifier after loading the DataFrames.jl package with
using DataFrames) types and functions are considered a part of the public API of the DataFrames.jl package. In general all such objects are documented in this manual (in case some documentation is missing please kindly report an issue here).
Breaking changes to public and documented API are avoided in DataFrames.jl where possible.
The following changes are not considered breaking:
- specific floating point values computed by operations may change at any time; users should rely only on approximate accuracy;
- in functions that use the default random number generator provided by Base Julia the specific random numbers computed may change across Julia versions;
- if the changed functionality is classified as a bug;
- if the changed behavior was not documented; two major cases are:
- in its implementation some function accepted a wider range of arguments that it was documented to handle - changes in handling of undocumented arguments are not considered as breaking;
- the type of the value returned by a function changes, but it still follows the contract specified in the documentation; for example if a function is documented to return a vector then changing its type from
PooledVectoris not considered as breaking;
- error behavior: code that threw an exception can change exception type thrown or stop throwing an exception;
- changes in display (how objects are printed);
- changes to the state of global objects from Base Julia whose state normally is considered volatile (e.g. state of global random number generator).
All types and functions that are part of public API are guaranteed to go through a deprecation period before a breaking change is made to them or they would be removed.
The standard practice is that breaking changes are implemented when a major release of DataFrames.jl is made (e.g. functionalities deprecated in a 1.x release would be changed in the 2.0 release).
In rare cases a breaking change might be introduced in a minor release. In such a case the changed behavior still goes through one minor release during which it is deprecated. The situations where such a breaking change might be allowed are (still such breaking changes will be avoided if possible):
- the affected functionality was previously clearly identified in the documentation as being subject to changes (for example in DataFrames.jl 1.4 release propagation rules of
:note-style metadata are documented as such);
- the change is on the border of being classified as a bug (in rare cases even if a behavior of some function was documented its consequences for certain argument combinations could be decided to be unintended and not wanted);
- the change is needed to adjust DataFrames.jl functionality to changes in Base Julia.
Please be warned that while Julia allows you to access internal functions or types of DataFrames.jl these can change without warning between versions of DataFrames.jl. In particular it is not safe to directly access fields of types that are a part of public API of the DataFrames.jl package using e.g. the
getfield function. Whenever some operation on fields of defined types is considered allowed an appropriate exported function should be used instead.
- Common API for types defined in DataFrames.jl