My focus are the algorithms not Jupyter. Jupyter is just a dev./presentation environment for me. I have a team of SW engineers who will take my algorithmic kernels and and “package" them within an app.
Thanks. Dipu Dipankar “Dipu” Ganguly [email protected] Cell: 408-203-8814 > On Aug 30, 2016, at 8:24 AM, Gabriel Pettier <[email protected]> > wrote: > > Python-for-android (from the Kivy project) is also an option > > You can keep your Python/CPython code, you just have to write the necessary > build recipes if you have any unusual C lib to build, and keep the rest of > your code logic as is, i don't think anybody tried to run Jupyter Notebook > with it, but there are bootstraps to display a web view instead of a kivy > app, so maybe it's totally possible to make ipynb run there and display the > interface in a normal android WebView. > > If what you want precisely is the algorithms you are devising, not > specifically IPython notebook, and you don't care for a semi-native look on > your app (kivy uses OpenGL to display everything, not the native toolkit, > which is unnacceptable for some people) you can build a different UI to > display your results with kivy, which is quite efficient for rapid > prototyping of applications. > > Regards > > On Tue, Aug 30, 2016 at 10:35:07PM +0800, Russell Keith-Magee wrote: >> Hi Dipu, >> >> The short answer is “not really” - especially if you’re hoping to have an >> actual user interface. >> >> The problem is that Android *really* wants you to run Java. You *can* >> compile CPython for Android, but that code won’t have any easy access to UI >> elements like buttons - you need to use JNI to bridge into the Java >> runtime. Another approach would be to play around with running your Python >> algorithm as a service on your tablet, which your UI then interacts with. >> >> Another approach is to use VOC, which is a Python compiler that outputs >> Java classfiles; this is a fairly new tool (so it isn’t feature complete >> with Python yet), and it isn’t the fastest option, so if you’re doing image >> processing, it might not be the best idea. >> >> Jython would also be an option - but Jython doesn’t currently compile on >> Android, due to some binary dependencies that Jython has. >> >> So - it *can* be done, but you’re going to need to get your hands dirty if >> you want to make it happen. There isn’t an easy, ready-to-use, >> just-deploy-it solution for Android. >> >> Yours, >> Russ Magee %-) >> >> On Tue, Aug 30, 2016 at 7:13 AM, Dipankar “Dipu” Ganguly <[email protected]> >> wrote: >> >>> I am developing image processing algorithms in IPython under Anaconda >>> Navigator’s Jupyter on my Mac running OS 10.11.6. Is there a way to convert >>> that code easily into an Android to run on an Android tablet? >>> >>> Thanks. >>> >>> Dipu >>> >>> >>> Dipankar Ganguly >>> Consultant: Strategy/Technology/Commercialization >>> Bothell, WA >>> Cell: 408-203-8814 >>> email: [email protected] >>> http://www.linkedin.com/in/dipugee >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> _______________________________________________ >>> Mobile-sig mailing list >>> [email protected] >>> https://mail.python.org/mailman/listinfo/mobile-sig >>> >>> > >> _______________________________________________ >> Mobile-sig mailing list >> [email protected] >> https://mail.python.org/mailman/listinfo/mobile-sig >
_______________________________________________ Mobile-sig mailing list [email protected] https://mail.python.org/mailman/listinfo/mobile-sig
