Ask Your Question
1

SageMathCell timeout issue

asked 2016-02-06 22:33:48 +0200

updated 2016-02-07 23:03:50 +0200

My website currently has live math using embedded Mathematica CDFs. Since Wolfram has told me that they are dropping support for the browser plugin, I would like to convert all of my live math to SageMathCell. In the first page I'm testing, I've found that evaluating the eigenvalues of a numerical 30x30 symmetric real matrix over RDF does not finish and the cell in which the evaluation runs becomes grayed out while the working indicator continues to spin endlessly.

Is this a limitation of your server or my local machine? Is there some way to keep the cell alive until the evaluation completes? Thanks.

edit retag flag offensive close merge delete

Comments

This is almost certainly a planned timeout - the service doesn't have endless computational capability. What is the matrix, though? (Update your question.) If it's symbolic then, yes, it will take forever because it happens in Maxima and solving such things is by Abel's Theorem not so easy :) but hopefully using the right base field should solve your problem.

kcrisman gravatar imagekcrisman ( 2016-02-07 00:41:35 +0200 )edit

(I could imagine that one could set up one's own cell server that allowed longer timeouts, or pay for such a service. Wolfram Alpha also only allows only so many computational resources before saying sayonara.)

kcrisman gravatar imagekcrisman ( 2016-02-07 00:42:23 +0200 )edit

Also, is the Wolfram support dropping only for Chrome? E.g. http://mathematica.stackexchange.com/... Just curious.

kcrisman gravatar imagekcrisman ( 2016-02-07 00:44:42 +0200 )edit

Finally, regarding http://analyticphysics.com/ you may be interested in http://mathbook.pugetsound.edu/

kcrisman gravatar imagekcrisman ( 2016-02-07 00:50:06 +0200 )edit

Question updated: matrix is real symmetric numerical over RDF. Are there any keywords that will speed up the diagonalization?

At some point I may set up my own server. For now I'll work with the known limitation.

paulmasson gravatar imagepaulmasson ( 2016-02-07 23:13:20 +0200 )edit

2 Answers

Sort by ยป oldest newest most voted
1

answered 2016-02-07 21:47:53 +0200

kcrisman gravatar image

According to the Sage cell admin, "We have 2 min CPU time limit per computation and 2 hours timeout for interacts/linked cells."

edit flag offensive delete link more
2

answered 2016-02-08 02:04:32 +0200

tmonteil gravatar image

I tried the following on sagecell, the answer came immediately

a = random_matrix(RDF,30,30)
a = a+a.transpose()
a.eigenvalues()

You can check at http://sagecell.sagemath.org/?z=eJxLV...

edit flag offensive delete link more

Comments

1

Nice test! I should have said that I'm performing the diagonalization 16 times then manipulating the results for further use, but I just ran your test 50 times on a 50x50 matrix with no problem. This gives me a baseline to figure out what causes the timeout.

paulmasson gravatar imagepaulmasson ( 2016-02-08 04:04:15 +0200 )edit

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

Question Tools

Stats

Asked: 2016-02-06 22:33:48 +0200

Seen: 1,052 times

Last updated: Feb 08 '16