[sage-support] How to make view open pdf-viewer in background
Hi there! I'm using view to display my graphs and posets. I would like to have it run the viewer (xdg-open, which in the case at hand is evince) in the background. This is on a linux box, running sage 6.5.beta2. Many many thanks! Martin -- You received this message because you are subscribed to the Google Groups sage-support group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-support+unsubscr...@googlegroups.com. To post to this group, send email to sage-support@googlegroups.com. Visit this group at http://groups.google.com/group/sage-support. For more options, visit https://groups.google.com/d/optout.
[sage-support] Re: Improving a Sage program that is heavy on matrix multipliaction
On 2014-12-07, Jernej azi.std...@gmail.com wrote:
--=_Part_3308_1073346028.1417954673282
Content-Type: multipart/alternative;
boundary==_Part_3309_733817265.1417954673282
--=_Part_3309_733817265.1417954673282
Content-Type: text/plain; charset=UTF-8
Just for fun I wrote an equivalent program in C and tested the Sage
function and an equivalent C program on 1 instance of the problem.
===
$ time ./a.out test
real0m0.366s
user0m0.306s
sys 0m0.060s
===
And Sage
===
$ time sage test.sage test
real2m6.285s
user2m5.256s
sys 0m0.858s
===
I thought that may be interesting to you as well.
Best,
Jernej
On Wednesday, 3 December 2014 22:33:46 UTC+1, Jernej wrote:
Dear sage-support,
I have stumbled into a performance bottleneck in one of my Sage programs.
I would like to share the relevant problem here in hope anyone has a
constructive suggestion for optimizing the given program.
I am given a n x n, (0,1) matrix C where n 20. C has up to 30% of
nonzero elements.
I define the inner product u,v = u (2I - C) v^t and need to compute the
set S of all (0,1) n-vectors u such that u,u = 2 and u,j = -1, where j
is the vector with all ones. Finally I need to record all pairs u,v from S
such that u,v == -1 or 0.
The one optimization that I use is to cache the product u*(2I-C). Other
than that I don't see any other way to optimize the program hence it looks
like
u,j=-1 is a linear condition on u.
Shouldn't you only pick up u from the corresponding hyperplane?
global vectors
field = RR # this looks like the fastest option
RDF should be faster, IMHO.
def init_vectors(n):
global vectors
vectors = [matrix(field,b) for b in product(*[[0,1]]*n)]
def foo(C,output):
global vectors
n = C.nrows()
# we compute the inverse in QQ just to gain a bit of numerical
stability
D = (2*identity_matrix(n)-C).inverse()
D = Matrix(field,D)
j = matrix(field, tuple([1 for _ in xrange(n)]))
cur = 1
vec2int = {}
cache = {}
for bm in vectors:
val = bm*D
if abs( (val*bm.transpose())[0,0]-2) = 0.01 and
abs((val*j.transpose())[0,0]+1) = 0.01:
vec2int[cur] = bm
output.write(str(el for el in bm[0])+'\n')
cache[cur] = val
cur+=1
for i in xrange(1, cur):
for j in xrange(i+1, cur):
iv = (cache[i]*vec2int[j].transpose())[0,0]
if abs(iv) = 0.001 or abs(iv+1) = 0.01:
output.write(str(i) + ' ' + str(j) + '\n')
===
For convenience's sake I have attached the result of %prun on one instance
of the problem. From the output of prun I suppose it would make sense to
cache the transpose of the vectors as well, though I am not sure this is
going to make the crucial difference.
On average it takes 200s to process one matrix and given that I have
millions of such matrices it would take years to process all the input.
Other than rewriting the whole thing in C I currently do not see any
viable solution. Hence I am wondering: Do you guys happen to see any clever
optimization? Is there a way to compute the named product more efficiently?
You could use Cython instead of C, and numpy arrays instead of Sage
matrices (numpy arrays interface quite well with Cython).
Alas, currently plain Sage matrices are very slow.
As your matrices are quite small, alternatively you might experiment
with computing modulo p, for p a prime or order bigger than n^2.
Best,
Jernej
--
You received this message because you are subscribed to the Google Groups
sage-support group.
To unsubscribe from this group and stop receiving emails from it, send an email
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-support.
For more options, visit https://groups.google.com/d/optout.
Re: [sage-support] Running Sage.app installed by another user
Then the chmod trick should work until we get a better fix. They workaround works fine. I chmoded 777 the executable /Applications/Sage.app/Contents/Resources/sage/sage. Thank you. exits normally, so I can’t use that. Maybe saving the user’s answer would be sufficient? So they would only get asked once (per sage version). Why not once per _location_ instead of version? The users home directory, their DOT_SAGE, is in their account and is persistent, while the .app version can change anytime between sessions. -Ivan Thanks a lot for your help, Jerome Tremblay LaCIM, UQAM -- You received this message because you are subscribed to the Google Groups sage-support group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-support+unsubscr...@googlegroups.com. To post to this group, send email to sage-support@googlegroups.com. Visit this group at http://groups.google.com/group/sage-support. For more options, visit https://groups.google.com/d/optout.
[sage-support] Re: error with help(notebook)
Il giorno martedì 9 dicembre 2014 02:33:50 UTC+1, kcrisman ha scritto: I would read the notebook help but it fails: sage: help (notebook) This is in fact expected. Try sage: notebook? for what you most likely want. Or, sage: import sagenb sage: help(sagenb.notebook) this is because `help` needs a Python module, not just a command. In general, the ? syntax is the way to go for help - see e.g. http://sagemath.org/doc/prep/Intro-Tutorial.html#help-inside-sage for more examples. Good luck! - kcrisman Thank you, just what i wanted. -- Informativa sulla Privacy: http://www.unibs.it/node/8155 -- You received this message because you are subscribed to the Google Groups sage-support group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-support+unsubscr...@googlegroups.com. To post to this group, send email to sage-support@googlegroups.com. Visit this group at http://groups.google.com/group/sage-support. For more options, visit https://groups.google.com/d/optout.
Re: [sage-support] Running Sage.app installed by another user
On Tue, Dec 9, 2014 at 8:37 AM, Jérôme Tremblay jerome.tremb...@gmail.com wrote: Then the chmod trick should work until we get a better fix. They workaround works fine. I chmoded 777 the executable /Applications/Sage.app/Contents/Resources/sage/sage. Thank you. Excellent. Glad to hear it. exits normally, so I can’t use that. Maybe saving the user’s answer would be sufficient? So they would only get asked once (per sage version). Why not once per _location_ instead of version? The users home directory, their DOT_SAGE, is in their account and is persistent, while the .app version can change anytime between sessions. Sorry, location is exactly what I meant. -Ivan -- You received this message because you are subscribed to the Google Groups sage-support group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-support+unsubscr...@googlegroups.com. To post to this group, send email to sage-support@googlegroups.com. Visit this group at http://groups.google.com/group/sage-support. For more options, visit https://groups.google.com/d/optout.
Re: [sage-support] Running Sage.app installed by another user
On Tue, Dec 9, 2014 at 8:03 AM, kcrisman kcris...@gmail.com wrote: Does someone want to open a trac ticket (or is there already one)? Go ahead, I still don't understand the details enough as to why it is giving that warning if they can indeed write to the file Sage will need to write to. Okay, http://trac.sagemath.org/ticket/17479 The reason is that Sage *will* write to .sage, but *might* have to write to the sage install because of sage-location (which they can't). Does that make sense? -Ivan -- You received this message because you are subscribed to the Google Groups sage-support group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-support+unsubscr...@googlegroups.com. To post to this group, send email to sage-support@googlegroups.com. Visit this group at http://groups.google.com/group/sage-support. For more options, visit https://groups.google.com/d/optout.
[sage-support]
Hello. Is there an easy way to load a picture into an array so as to apply a matrix transformation to all the xy-corrdinates ? If a solution without using numpy exists, this would be cool. *Christophe BAL* *Enseignant de mathématiques en Lycée **et développeur Python amateur* *---* *French math teacher in a Lycée **and **Python **amateur developer* -- You received this message because you are subscribed to the Google Groups sage-support group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-support+unsubscr...@googlegroups.com. To post to this group, send email to sage-support@googlegroups.com. Visit this group at http://groups.google.com/group/sage-support. For more options, visit https://groups.google.com/d/optout.
Re: [sage-support] Running Sage.app installed by another user
The reason is that Sage *will* write to .sage, but *might* have to write to the sage install because of sage-location (which they can't). Oh right, that's right. -- You received this message because you are subscribed to the Google Groups sage-support group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-support+unsubscr...@googlegroups.com. To post to this group, send email to sage-support@googlegroups.com. Visit this group at http://groups.google.com/group/sage-support. For more options, visit https://groups.google.com/d/optout.
Re: [sage-support]
Hello, I would say PIL is an alternative (but less powerful than numpy). A bit of googling brought me to this very nice post http://stackoverflow.com/questions/14177744/how-does-perspective-transformation-work-in-pil Best Vincent 2014-12-10 0:33 UTC+01:00, Christophe Bal projet...@gmail.com: Hello. Is there an easy way to load a picture into an array so as to apply a matrix transformation to all the xy-corrdinates ? If a solution without using numpy exists, this would be cool. *Christophe BAL* *Enseignant de mathématiques en Lycée **et développeur Python amateur* *---* *French math teacher in a Lycée **and **Python **amateur developer* -- You received this message because you are subscribed to the Google Groups sage-support group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-support+unsubscr...@googlegroups.com. To post to this group, send email to sage-support@googlegroups.com. Visit this group at http://groups.google.com/group/sage-support. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups sage-support group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-support+unsubscr...@googlegroups.com. To post to this group, send email to sage-support@googlegroups.com. Visit this group at http://groups.google.com/group/sage-support. For more options, visit https://groups.google.com/d/optout.
