[sage-support] Re: semidefinite programming using csdp -- python import errors

2013-02-15 Thread Dima Pasechnik
I've put these little changes on https://github.com/dimpase/pycsdp and https://github.com/dimpase/csdp This works with Sage 5.6 on OSX 10.6.8 On Saturday, 24 March 2012 04:31:05 UTC+8, Dima Pasechnik wrote: > > On 2012-03-23, Dima Pasechnik <> wrote: > > On 2012-03-23, Robert Samal <> wrote: >

[sage-support] Re: semidefinite programming using csdp -- python import errors

2012-04-07 Thread Robert Samal
Thanks a lot, Dmitrii! It works for me now. (Actually for quite some time, but now I can happily report, that compared to cvxopt that I was using before, using csdp is about 10-times faster!) I didn't know about sage -sh , that is a very useful trick (perhaps it should be documented more, or

Re: [sage-support] Re: semidefinite programming using csdp -- python import errors

2012-03-27 Thread Emil
>> Precision could be lost in >> the double->string->double translations (?). > why would you think there is such a translation? I'm referring to using the stand-alone CSDP. You have to write your matrices, and read the solution matrix, as a text file. pycsdp avoids this, as would a Cython interfa

[sage-support] Re: semidefinite programming using csdp -- python import errors

2012-03-26 Thread Dima Pasechnik
On 2012-03-26, Emil wrote: > On 26 March 2012 17:37, Dima Pasechnik wrote: >> On 2012-03-25, Emil wrote: >>> I've also been using CSDP from Sage lately. I am calling the >>> stand-alone program, rather than using the Python interface (which >>> isn't a very good solution). >> why?  Is it buggy?

Re: [sage-support] Re: semidefinite programming using csdp -- python import errors

2012-03-26 Thread Emil
On 26 March 2012 17:37, Dima Pasechnik wrote: > On 2012-03-25, Emil wrote: >> I've also been using CSDP from Sage lately. I am calling the >> stand-alone program, rather than using the Python interface (which >> isn't a very good solution). > why?  Is it buggy? Well, you can't stop/pause/save in

[sage-support] Re: semidefinite programming using csdp -- python import errors

2012-03-26 Thread Dima Pasechnik
On 2012-03-25, Emil wrote: > I've also been using CSDP from Sage lately. I am calling the > stand-alone program, rather than using the Python interface (which > isn't a very good solution). why? Is it buggy? > > I was wondering whether using pycsdp is the right way to go though. > Wouldn't it be

Re: [sage-support] Re: semidefinite programming using csdp -- python import errors

2012-03-25 Thread Emil
I've also been using CSDP from Sage lately. I am calling the stand-alone program, rather than using the Python interface (which isn't a very good solution). I was wondering whether using pycsdp is the right way to go though. Wouldn't it be best to write a small Cython wrapper for the C library? (I

[sage-support] Re: semidefinite programming using csdp -- python import errors

2012-03-24 Thread Dima Pasechnik
On 2012-03-24, Dima Pasechnik wrote: > On 2012-03-23, Dima Pasechnik wrote: > [...] >> >> I have had success with installing and running this extenstion on MacOSX >> 10.6.8, under Sage 4.8. (That is, I didn't even use Atlas - Sage uses >> native MacOSX lapack/blas) >> >> >> Instructions: >> fire

[sage-support] Re: semidefinite programming using csdp -- python import errors

2012-03-23 Thread Dima Pasechnik
On 2012-03-23, Dima Pasechnik wrote: [...] > > I have had success with installing and running this extenstion on MacOSX > 10.6.8, under Sage 4.8. (That is, I didn't even use Atlas - Sage uses > native MacOSX lapack/blas) > > > Instructions: > fire up Sage shell: sage -sh > 1) Compile Csdp static l

[sage-support] Re: semidefinite programming using csdp -- python import errors

2012-03-23 Thread Dima Pasechnik
On 2012-03-23, Dima Pasechnik wrote: > On 2012-03-23, Robert Samal wrote: >> --=_Part_772_24642055.1332490982625 >> Content-Type: multipart/alternative; >> boundary="=_Part_773_30231439.1332490982625" >> >> --=_Part_773_30231439.1332490982625 >> Content-Type: text/plain; charset

[sage-support] Re: semidefinite programming using csdp -- python import errors

2012-03-23 Thread Dima Pasechnik
On 2012-03-23, Robert Samal wrote: > --=_Part_772_24642055.1332490982625 > Content-Type: multipart/alternative; > boundary="=_Part_773_30231439.1332490982625" > > --=_Part_773_30231439.1332490982625 > Content-Type: text/plain; charset=ISO-8859-1 > > >> >> Can you post the logs f

[sage-support] Re: semidefinite programming using csdp -- python import errors

2012-03-23 Thread Dima Pasechnik
On 2012-03-23, Jason Grout wrote: > On 3/22/12 8:58 PM, Robert Samal wrote: >> >> Hi, >> >> I'm trying to use library Csdp in >> Sage by means of Python numpy wrapper for CSDP called pycsdp (by B. >> Kern) >>

[sage-support] Re: semidefinite programming using csdp -- python import errors

2012-03-23 Thread Robert Samal
> > Can you post the logs from doing sage -python setup.py instasll, as well > as explicit instructions (where to download, etc.) for how to duplicate > the error? > Sure. It's rather long, but I suppose if it works it would be useful for other people, too. I was hoping someone could spot the p

[sage-support] Re: semidefinite programming using csdp -- python import errors

2012-03-22 Thread Jason Grout
On 3/22/12 8:58 PM, Robert Samal wrote: Hi, I'm trying to use library Csdp in Sage by means of Python numpy wrapper for CSDP called pycsdp (by B. Kern) . I was assuming that it'll be