Mayavi: 3D visualization of scientific data in Python
Mayavi docs: http://docs.enthought.com/mayavi/mayavi/ TVTK docs: http://docs.enthought.com/mayavi/tvtk
Vision
Mayavi seeks to provide easy and interactive visualization of 3D data. It does this by the following:
- an (optional) rich user interface with dialogs to interact with all data and objects in the visualization.
- a simple and clean scripting interface in Python, including one-liners, a-la mlab, or object-oriented programming interface.
- harnesses the power of the VTK toolkit without forcing you to learn it.
Additionally Mayavi strives to be a reusable tool that can be embedded in your applications in different ways or combined with the envisage application-building framework to assemble domain-specific tools.
Mayavi is part of the Enthought Tool Suite (ETS).
Features
Mayavi is a general purpose, cross-platform tool for 2-D and 3-D scientific data visualization. Its features include:
- Visualization of scalar, vector and tensor data in 2 and 3 dimensions
- Easy scriptability using Python
- Easy extendability via custom sources, modules, and data filters
- Reading several file formats: VTK (legacy and XML), PLOT3D, etc.
- Saving of visualizations
- Saving rendered visualization in a variety of image formats
- Convenient functionality for rapid scientific plotting via mlab (see mlab documentation)
- See the Mayavi Users Guide for more information.
Unlike its predecessor MayaVi1, Mayavi has been designed with scriptability and extensibility in mind from the ground up. While the mayavi2 application is usable by itself, it may be used as an Envisage plugin which allows it to be embedded in user applications natively. Alternatively, it may be used as a visualization engine for any application.
Quick start
If you are new to mayavi it is a good idea to read the online user manual which should introduce you to how to install and use it.
If you have installed mayavi as described in the next section, you should be able to launch the mayavi2 application and also run any of the examples in the examples directory.
Installation
By itself Mayavi is not a difficult package to install but its dependencies are unfortunately rather heavy. However, many of these dependencies are now available as wheels on PyPI. The two critical dependencies are,
The latest VTK wheels are available on all the major platforms (Windows, MacOS, and Linux), but only for 64 bit machines. Python 3.x is fully supported on all these operating systems and Python 2.7.x on MacOS and Linux. If you are out of luck, and your platform is not supported then you will need to install VTK yourself using your particular distribution as discussed in the General Build and Installation instructions
On Python 3.x you will need to install PyQt5 and wheels are available for this. On 2.7.x you have more options, and can use PySide, PyQt4, and wxPython. These can be installed from pip or from your package manager.
Currently, Mayavi itself should work with the new wxPython 4.x. However, traitsui, pyface, and other ETS packages do not yet support it so the UI will not work correctly. Older versions should work. PyQt/PySide/PySide2 should work largely out of the box.
Latest stable release
As of the latest release, i.e. 4.6.0 and above, if you are using Python 3.x and are on a 64 bit machine, installation via pip is the easiest and is as follows:
$ pip install mayavi $ pip install PyQt5
Thats it!
If you are unable to do this, read the documentation above and find a way to install VTK and a suitable UI toolkit and then repeat the above.
If you are interested in the jupyter notebook support as well, do the following (after ensuring that you have jupyter installed of course):
$ jupyter nbextension install --py mayavi --user $ jupyter nbextension enable --py mayavi --user
You will also need to have ipywidgets and ipyevents installed. These can be installed via pip or your favorite package manager.
Bleeding edge
If you want to install the latest version of Mayavi from github, you can simply do the following:
$ git clone https://github.com/enthought/mayavi.git $ cd mayavi $ pip install -r requirements.txt $ pip install PyQt5 # replace this with any supported toolkit $ python setup.py install # or develop
Add the jupyter nbextensions using the instructions in the section above and you should be good to go.
Documentation
More documentation is available in the online user manual or in docs
directory of the sources. This includes a man page for the mayavi2
application, a users guide in HTML and PDF format and documentation for
mlab.
More documentation in the form of workshop/tutorial material is available here:
- https://github.com/prabhuramachandran/mayavi-tutorial
- https://github.com/prabhuramachandran/mayavi-workshop
Tutorial Videos
Here are some tutorial videos that you can watch to learn Mayavi:
- SciPy 2018 Mayavi tutorial (3 hrs):
Examples
Examples are all in the examples
directory of the source or the git clone.
The docs and examples do not ship with the binary eggs. The examples directory
also contains some sample data.
Test suite
The basic test suites for tvtk and mayavi can be run using nose:
nosetests -v tvtk/tests nosetests -v mayavi
The integration tests:
cd integrationtests/mayavi python run.py
Bug tracker, mailing list etc.
The bug tracker is available in github Please provide info and details on platform, python, vtk and gui backends and their versions. If possible, a small example replicating the the problem.
If you have questions you could ask on the Mayavi-users mailing list. This is used by some folks and is not too active. Another mailing list that may be of use is the ETS Users mailing list. This is a more general list where a lot of folks experienced with the Enthought Tool Suite are available.
Authors and Contributors
Core contributors:
Prabhu Ramachandran: primary author.
Previous contributors:
Gaël Varoquaux: mlab, icons, many general improvements and maintenance.
Deepak Surti: Upgrade to VTK 5.10.1, VTK 6.x with new pipeline.
Support and code contributions from Enthought Inc.
Patches from many people (see the release notes), including K K Rai and R A Ambareesha for tensor support, parametric source and image data.
Many thanks to all those who have submitted bug reports and suggestions for further enhancements.