Improved cl.oquence documentation
1 view
Skip to first unread message
Cyrus Omar
Jun 1, 2010, 6:51:29 PM6/1/10
to ahh-d...@googlegroups.com
Hey,
There is much improved cl.oquence documentation (in addition to the previously mentioned ahh.cl and general documentation) now -- http://ahh.bitbucket.org/.
Next up is writing some test cases, then merging the cl.oquence branch with the main branch (which might break some of your code in minor ways, Michael, sorry), then addressing the various unimplemented features listed in the documentation on an as-needed basis. As-needed meaning as I need them to implement spiking networks or as any of you need them to implement whatever it is you're doing.
Please don't hesitate to use this list to communicate, everything is still in a fairly early stage of testing and completion so its unlikely you'll be able to be autonomous just yet.
Before my DOE conference in 2.5 weeks I hope to have ahh.cl and cl.oquence at the 1.0 release candidate stage based on work getting spiking networks to work at at least the level of functionality the current CUDA implementation works. By the end of the summer I'm hoping to get a substantial set of people on the codebase.
Cyrus
There is much improved cl.oquence documentation (in addition to the previously mentioned ahh.cl and general documentation) now -- http://ahh.bitbucket.org/.
Next up is writing some test cases, then merging the cl.oquence branch with the main branch (which might break some of your code in minor ways, Michael, sorry), then addressing the various unimplemented features listed in the documentation on an as-needed basis. As-needed meaning as I need them to implement spiking networks or as any of you need them to implement whatever it is you're doing.
Please don't hesitate to use this list to communicate, everything is still in a fairly early stage of testing and completion so its unlikely you'll be able to be autonomous just yet.
Before my DOE conference in 2.5 weeks I hope to have ahh.cl and cl.oquence at the 1.0 release candidate stage based on work getting spiking networks to work at at least the level of functionality the current CUDA implementation works. By the end of the summer I'm hoping to get a substantial set of people on the codebase.
Cyrus
Reply all
Reply to author
Forward
0 new messages