Re: [sage-devel] Problems in the computation of hyperbolicity.

2014-04-22 Thread Vincent Delecroix
Hello again, I was a bit wrong with my computations. Have a look at the file SAGE_ROOT/src/sage/graphs/distance_all_pairs.pyx You will see that matrices for distances are implemented as unsigned short (whose maximum is 65535=2^16) while eccentricity is made of int (whose maximum depends on your a

Re: [sage-devel] Problems in the computation of hyperbolicity.

2014-04-22 Thread Vincent Delecroix
Hello Miguel, If you need support in Sage you should use the sage-support googlegroups or ask.sagemath.org. The sage-devel mailing list is about development and bug report. To answer your question, if you want to compute all distances in a graph with more than 65535 vertices then your RAM must be

[sage-devel] Problems in the computation of hyperbolicity.

2014-04-22 Thread Miguel Camelo
Hi everyone, I'm interesting in to compute both the diameter and hyperbolicity of large graphs (between 10k and 10M of nodes) using SAGE. At the beginning I had problems computing the diameter. However I solved the problem using the solutions presented in the ticked #15507. Then, I tried to co