I'm looking into using SMILE C++ libraries on Angstrom Linux. Specifically, I'm thinking of using Angstrom Linux port of SMILE for an ARM Cortex v7 processor on a BeagleBoard-xM. I think Angstrom’s cross-compiler tool-chain can be used to compile the library. However, I could not find the sources. Where do I download the SMILE sources from? I would like to contribute the ARMv7 pre-built libraries back to the community.
-- Sumant
SMILE source code
Re: SMILE source code
Hi Sutambe,
SMILE is not open source software.
I'll contact you over email to see if we can work out a way to compile it for your platform.
Best,
Martijn
(PhD Student, Decision Systems Lab)
SMILE is not open source software.
I'll contact you over email to see if we can work out a way to compile it for your platform.
Best,
Martijn
(PhD Student, Decision Systems Lab)
-
peter.serwylo
- Posts: 1
- Joined: Fri Mar 30, 2012 3:03 am
Re: SMILE source code
Hey Martijn,
Has the DSL considered open sourcing Genie and Smile? I find Genie a great piece of software, and I'm sure many of your users would be happy to coontribute to making it better (including myself). Looking at the changelog (http://genie.sis.pitt.edu/download/changelog.txt), I can see that it is still being developed, but history shows that you would probably have many more features and bug fixes if it was under an appropriate open source license.
Seeing as there are "no strings attached" when you download the software, other than acknowledgements, is there specific arguments against the open source route in your lab? For the record, I completely understand if there is some reason behind it, I was just curious if it has been considered.
cheers,
pete.
Has the DSL considered open sourcing Genie and Smile? I find Genie a great piece of software, and I'm sure many of your users would be happy to coontribute to making it better (including myself). Looking at the changelog (http://genie.sis.pitt.edu/download/changelog.txt), I can see that it is still being developed, but history shows that you would probably have many more features and bug fixes if it was under an appropriate open source license.
Seeing as there are "no strings attached" when you download the software, other than acknowledgements, is there specific arguments against the open source route in your lab? For the record, I completely understand if there is some reason behind it, I was just curious if it has been considered.
cheers,
pete.
-
marek [BayesFusion]
- Site Admin
- Posts: 438
- Joined: Tue Dec 11, 2007 4:24 pm
Re: SMILE source code
Pete,
We have considered making GeNIe and SMILE open source and discussed internally arguments for and against, doing some research as well. We have finally decided against it. So, compiled version only
.
Cheers,
Marek
We have considered making GeNIe and SMILE open source and discussed internally arguments for and against, doing some research as well. We have finally decided against it. So, compiled version only
.Cheers,
Marek
Re: SMILE source code
I pretty sure that in this case this was not a good decision. Perhaps 10 years go that made sense but nowadays SMILE is developing too slow. I would be glad if I could help you out with MacOSX binaries but I simply can't and you cannot afford a Mac... I've been working with SMILE for 2 years now and I've got a ton of improvements I could possibly share. I even rewrote jSMILE wrapper and fixed some issue's with SMILE .NET.
Kind regards,
Piotr
Kind regards,
Piotr
-
marek [BayesFusion]
- Site Admin
- Posts: 438
- Joined: Tue Dec 11, 2007 4:24 pm
Re: SMILE source code
Interesting . I do decision support for a living and I assure you that calling somebody else's decision bad is a pretty bold act. The reason for that is that you may have no idea of that person's subjective estimates of uncertainty and may have no clue of his/her utility. I assure you that we have thought about this decision carefully and also more recently than 10 years ago. Also, SMILE and GeNIe have been developing rapidly, so please allow me to disagree with your judgment. You are wrong about Mac as well, I'm afraid. I think that you are not being very nice. I propose that we terminate this direction of discussion, OK?
This having said, thanks for your contributions! The sources of SMILE.NET and jSMILE wrappers are available from our web site. I assume that you have shared your improvements with other users? I also invite you to contribute to the documentation -- it is in Wiki format and every user can add to it.
Cheers,
Marek
. I do decision support for a living and I assure you that calling somebody else's decision bad is a pretty bold act. The reason for that is that you may have no idea of that person's subjective estimates of uncertainty and may have no clue of his/her utility. I assure you that we have thought about this decision carefully and also more recently than 10 years ago. Also, SMILE and GeNIe have been developing rapidly, so please allow me to disagree with your judgment. You are wrong about Mac as well, I'm afraid. I think that you are not being very nice. I propose that we terminate this direction of discussion, OK?This having said, thanks for your contributions! The sources of SMILE.NET and jSMILE wrappers are available from our web site. I assume that you have shared your improvements with other users? I also invite you to contribute to the documentation -- it is in Wiki format and every user can add to it.
Cheers,
Marek
Re: SMILE source code
Firstly, sorry. I didn't inted to offend you (or anyone involved in SMILE) nor I wanted to sound rude. I hope that my poor language can be a bit of explanation. I really appreciate your work and I am a huge fan of SMILE.
Want I wanted to say is that I really believe that SMILE could develop faster if it was an open-source project. I don't know if it fits your business model. I just know that there are lots of clever developers and perhaps remarkable companies that would be glad to contribute to this great project.
Yes, I've shared my most important contributions with others - you can track my posts and see that I continously report my hints and issues on SMILE.
Concerning MacOS, just like I've reported a month ago, there are no new binaries for MacOS so you cannot compile newest jSMILE which has changed. shooltz told me that you don't have a Mac right now and you cannot compile SMILE for this platform. I *really* needed these binaries, so I offered my help. I admit that I shouldn't say that you cannot afford Mac - and sorry again - but I just wanted to express that I won't see MacOS binaries soon (if ever). Having a Mac I need to run SMILE on Windows virtual machine.
Kind regards and Happy Easter,
Piotr
Want I wanted to say is that I really believe that SMILE could develop faster if it was an open-source project. I don't know if it fits your business model. I just know that there are lots of clever developers and perhaps remarkable companies that would be glad to contribute to this great project.
Yes, I've shared my most important contributions with others - you can track my posts and see that I continously report my hints and issues on SMILE.
Concerning MacOS, just like I've reported a month ago, there are no new binaries for MacOS so you cannot compile newest jSMILE which has changed. shooltz told me that you don't have a Mac right now and you cannot compile SMILE for this platform. I *really* needed these binaries, so I offered my help. I admit that I shouldn't say that you cannot afford Mac - and sorry again - but I just wanted to express that I won't see MacOS binaries soon (if ever). Having a Mac I need to run SMILE on Windows virtual machine.
Kind regards and Happy Easter,
Piotr
-
marek [BayesFusion]
- Site Admin
- Posts: 438
- Joined: Tue Dec 11, 2007 4:24 pm
Re: SMILE source code
Piotr,
I do not feel offended and practically never do until I make sure that somobody really intends to offend me.
Thanks a lot for being an active and very socially conscious user. I have seen many of your posts and I'd like to thank you for them.
Newest Mac compile is coming but having a Mac is not the issue -- we do have a Mac but are waiting for an opening the firewall at the University of Pittsburgh for Tomek (he is telecommuting) to be able to access it. If a Mac compile is very urgent for you, I can ask somebody in the lab locally to perform it. The current Mac binaries on the web are indeed a few years old. This reflects the fact that Mac users are still a small minority. This can, of course, change. The old Mac binaries are stable but are lacking the newest developments. Do you need SMILE of jSMILE for the Mac? I assume SMILE is enough for you, as you can compile the jSMILE sources with SMILE locally, correct? We can continue through regular Email if you want. My Email is marek@sis.pitt.edu.
Cheers,
Marek
I do not feel offended and practically never do until I make sure that somobody really intends to offend me
.Thanks a lot for being an active and very socially conscious user. I have seen many of your posts and I'd like to thank you for them.
Newest Mac compile is coming but having a Mac is not the issue -- we do have a Mac but are waiting for an opening the firewall at the University of Pittsburgh for Tomek (he is telecommuting) to be able to access it. If a Mac compile is very urgent for you, I can ask somebody in the lab locally to perform it. The current Mac binaries on the web are indeed a few years old. This reflects the fact that Mac users are still a small minority. This can, of course, change. The old Mac binaries are stable but are lacking the newest developments. Do you need SMILE of jSMILE for the Mac? I assume SMILE is enough for you, as you can compile the jSMILE sources with SMILE locally, correct? We can continue through regular Email if you want. My Email is marek@sis.pitt.edu.
Cheers,
Marek
Re: SMILE source code
Thank you Marek, I'll contact you via e-mail.