Some source files and icons may be under other authorship/licenses; see NOTICE.txt.

Overview

Spyder is a powerful scientific environment written in Python, for Python, and designed by and for scientists, engineers and data analysts. It offers a unique combination of the advanced editing, analysis, debugging, and profiling functionality of a comprehensive development tool with the data exploration, interactive execution, deep inspection, and beautiful visualization capabilities of a scientific package.

Beyond its many built-in features, its abilities can be extended even further via its plugin system and API. Furthermore, Spyder can also be used as a PyQt5 extension library, allowing you to build upon its functionality and embed its components, such as the interactive console, in your own software.

For more general information about Spyder and to stay up to date on the latest Spyder news and information, please check out our new website.

Core components

Editor

Work efficiently in a multi-language editor with a function/class browser, real-time code analysis tools (pyflakes, pylint, and pycodestyle), automatic code completion (jedi and rope), horizontal/vertical splitting, and go-to-definition.
Interactive console

Harness the power of as many IPython consoles as you like with full workspace and debugging support, all within the flexibility of a full GUI interface. Instantly run your code by line, cell, or file, and render plots right inline with the output or in interactive windows.
Documentation viewer

Render documentation in real-time with Sphinx for any class or function, whether external or user-created, from either the Editor or a Console.
Variable explorer

Inspect any variables, functions or objects created during your session. Editing and interaction is supported with many common types, including numeric/strings/bools, Python lists/tuples/dictionaries, dates/timedeltas, Numpy arrays, Pandas index/series/dataframes, PIL/Pillow images, and more.
Development tools

Examine your code with the static analyzer, trace its execution with the interactive debugger, and unleash its performance with the profiler. Keep things organized with project support and a builtin file explorer, and use find in files to search across entire projects with full regex support.

Documentation

You can read the Spyder documentation online on the Spyder Docs website.

Installation

For a detailed guide to installing Spyder, please refer to our installation instructions.

The easiest way to install Spyder on any of our supported platforms is to download it as part of the Anaconda distribution, and use the conda package and environment manager to keep it and your other packages installed and up to date.

If in doubt, you should always install Spyder via this method to avoid unexpected issues we are unable to help you with; it generally has the least likelihood of potential pitfalls for non-experts, and we may be able to provide limited assistance if you do run into trouble.

Other install options exist, including:

The WinPython distribution for Windows
The MacPorts project for macOS
Your distribution's package manager (i.e. apt-get, yum, etc) on Linux
The pip package manager, included with most Python installations

However, we lack the resources to provide individual support for users who install via these methods, and they may be out of date or contain bugs outside our control, so we recommend the Anaconda version instead if you run into issues.

Troubleshooting

Before posting a report, please carefully read our Troubleshooting Guide and search the issue tracker for your error message and problem description, as the great majority of bugs are either duplicates, or can be fixed on the user side with a few easy steps. Thanks!

Contributing and Credits

Spyder was originally created by Pierre Raybaut, and is currently maintained by Carlos Córdoba and an international community of volunteers.

You can join us—everyone is welcome to help with Spyder! Please read our contributing instructions to get started!

Certain source files are distributed under other compatible permissive licenses and/or originally by other authors. The icons for the Spyder 3 theme are derived from Font Awesome 4.7 (© 2016 David Gandy; SIL OFL 1.1). Most Spyder 2 theme icons are sourced from the Crystal Project icon set (© 2006-2007 Everaldo Coelho; LGPL 2.1+). Other Spyder 2 icons are from Yusuke Kamiyamane (© 2013 Yusuke Kamiyamane; CC-BY 3.0), the FamFamFam Silk icon set (© 2006 Mark James; CC-BY 2.5), and the KDE Oxygen icons (© 2007 KDE Artists; LGPL 3.0+).

See NOTICE.txt for full legal information.

Running from a Github clone

Spyder can be run directly from the source code, hosted on the Spyder github repo. You may want to do this for fixing bugs in Spyder, adding new features, learning how Spyder works or to try out development versions before they are officially released.

If using conda (strongly recommended), this can be done by running the following from the command line (the Anaconda Prompt, if on Windows):

conda install spyder
conda remove spyder
git clone https://github.com/spyder-ide/spyder.git
cd spyder
python bootstrap.py

You also need to make sure the correct spyder-kernels version is installed for the version of Spyder you are testing. The above procedure will give you spyder-kernels 0.x for the 3.x branch (Spyder 3), so to run the master branch (Spyder 4) you need to additionally execute:

conda install -c spyder-ide spyder-kernels=1.*

Alternatively, you can use pip to install PyQt5 and the other runtime dependencies listed below. However, beware: this method is recommended for experts only, and you'll need to solve any problems on your own.

See the installation instructions for more details on all of this.

Dependencies

Important Note: Most or all of the dependencies listed below come with Anaconda and other scientific Python distributions, so you don't need to install them separately in those cases.

Build dependencies

When installing Spyder from its source package, the only requirement is to have a Python version greater than 2.7 or 3.4 (Python <=3.3 is no longer supported).

Runtime dependencies

Python 2.7 or 3.4+: The core language Spyder is written in and for.
PyQt5 5.5+: Python bindings for Qt, used for Spyder's GUI.
qtconsole 4.2.0+: Enhanced Python interpreter.
Rope 0.9.4+ and Jedi 0.9.0+: Editor code completion, calltips and go-to-definition.
Pyflakes: Real-time code analysis.
Sphinx: Rich text mode for the Help pane.
Pygments 2.0+: Syntax highlighting for all file types it supports.
Pylint: Static code analysis.
Pycodestyle: Real-time code style analysis.
Psutil: CPU and memory usage on the status bar.
Nbconvert: Manipulation of notebooks in the Editor.
Qtawesome 0.5.7+: To have an icon theme based on FontAwesome.
Pickleshare: Show import completions on the Python consoles.
PyZMQ: Run introspection services asynchronously.
QtPy 1.5.0+: Abstraction layer for Python Qt bindings so that Spyder can run on multiple Qt bindings and versions.
Chardet: Character encoding auto-detection in Python.
Numpydoc: Used by Jedi to get function return types from Numpydocstrings.
Cloudpickle: Serialize variables in the IPython kernel to send to Spyder.
spyder-kernels 0.4.3+;1.0-: Jupyter kernels for the Spyder console.
keyring: Save Github credentials to report errors securely.