You also might want to consider licensing of these frameworks. I am a big fan of open source myself, and think that open source is the way the science has to be done, but your institution might think the other way.
To break it down:
PyQt is GPL, so that any code simply using it MUST also be released under compatible license (there is a commercial version though). PySide is LGPL, so that if you only link to it and not modify it you are not bound to release your program as open source. All other toolkits are BSD-like licensed, so no problems there too.
I have personnaly used wxPython, PySide and a little bit of Tk. Tk looks ugly on all platforms, wx and Qt are both nice and powerful, although wx feels a bit more "pythonic" to me. As for the footprint, if you only include those parts of Qt that your really need the footprint is not that different.
Answering this is impossible without first knowing what your requirements/wishes are.
Should it be:
-available in default python installs (Tkinter is the obvious choice then)
-possible for a user to apply 'themes' (PyQT/PySide or PyGTK jump to mind in this case)
-similar to a GUI widget set you already know from another language? (fill in your own preference in this case)
-have a small footprint (don't use PyQT/PySide then, because it fairly large)
-etc.
For some more details, see for example this page:
http://pythoncentral.org/introduction-python-gui-development/
In addition to the help already provided by Uriel and Jos:
Did you consider PyGObject as the successor to PyGTK, which is seemingly being phased out?
Hi,
If you want to create GUI with plots, I would recommend PyQT with the additionnal packages pyqwt. I found it quite easy to use and gave pretty good GUI.
You should also consider the guiplot package which is included with pythonxy (a excellent scientific oriented python distribution by the way, https://code.google.com/p/pythonxy/)
Happy coding
You also might want to consider licensing of these frameworks. I am a big fan of open source myself, and think that open source is the way the science has to be done, but your institution might think the other way.
To break it down:
PyQt is GPL, so that any code simply using it MUST also be released under compatible license (there is a commercial version though). PySide is LGPL, so that if you only link to it and not modify it you are not bound to release your program as open source. All other toolkits are BSD-like licensed, so no problems there too.
I have personnaly used wxPython, PySide and a little bit of Tk. Tk looks ugly on all platforms, wx and Qt are both nice and powerful, although wx feels a bit more "pythonic" to me. As for the footprint, if you only include those parts of Qt that your really need the footprint is not that different.
I've used PyQt for academic and commercial software. In both cases, I had good results.
I prefer PyQt4/Pyside. If you don't need plots updated in "real-time", you can embed matplotlib plots inside both PyQt4 and Pyside. If you need rapid updating of plots, then I recommend using pyqtgraph inside either one. I have been using pyqtgraph for almost all GUI programs just because I like a lot of the features it has.
I prefer Qt based framework (PyQt4, PyQt5 or Pyside). You can also used this for developt commercial software.
I use Kivy which is a cross-platform framework with the ability to create applications working in all platforms. I prepared a book titled "
Building Android Apps in Python Using Kivy with Android Studio" available here https://www.apress.com/gp/book/9781484250303 . It is a beginners guide to build Android apps in Kivy and distribute the APKs to Google Play as regular Android apps.
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Programming Languages
- Python