Changes between Initial Version and Version 1 of Ticket #319
- Timestamp:
- 10/06/09 12:28:10 (15 years ago)
Legend:
- Unmodified
- Added
- Removed
- Modified
-
Ticket #319
-
Property
Status
changed from
new
toassigned
-
Property
Summary
changed from
Infinite capacities in push-relabel algorithms
toInfinite capacities in Preflow
-
Property
Status
changed from
-
Ticket #319 – Description
initial v1 1 If you run a push-relabel algorithm using floating point capacity type on a graph having some arcs of infinite capacity (represented by either the `infinity()` or the `max()` value of the type), then it could cause problems, since the algorithm usually saturate arcs.1 If you run `Preflow` using floating point capacity type on a network in which the source node has some outgoing arcs of infinite capacity (represented by either the `infinity()` or the `max()` value of the type), then it will cause problems, since the algorithm saturates these arcs. 2 2 3 To prevent this problem, we could replace the infinite capacities with sufficiently large finite values. For `Preflow` it would require the finding a finite value cut (if one exists). However for `Circulation` the `SUM[-sup(v) : sup(v)<0]` value could be used (if it is finite), so it would be simpler.3 To prevent this problem, we could replace the infinite capacities with sufficiently large finite values. However it would require the finding a finite value cut (if one exists). 4 4 5 I suggest to either make the se algorithms safer in this manner or add a warning to the doc of these classes and the dimacs solver about this problem.5 I suggest to either make the algorithm safer in this manner or add a warning about this problem to the doc of the class and the dimacs solver.