A heuristic tree decomposition algorithm. First a tree decomposition is build using vertex elimination orderings, where an addressable priority queue determines the order based on a greedy approach.
For non-simplicial vertices the key value is given by edge density times bagsize()-1/bagsize
elimination [options] <.td file> <vertex>
Argument | Description | Required |
---|---|---|
<.graph file> |
Path to input graph file | yes |
<vertex> |
Provides an element of randomness for a key | yes |
Option | Description | Default |
---|---|---|
--print_td=true|false |
Print full tree decomposition | false |
data/test_set/mytest.graph
7 9 10
10 2 7
9 1 3 4
8 2 6 7
7 2 5
6 4 6
5 3 5 7
4 1 3 6
data/test_set/mytest.td
s td 7 4 7
b 1 1 2 7
b 2
b 3
b 4
b 5 2 4 5
b 6 2 5 6
b 7 2 3 6 7
1 7
2 3
2 4
4 5
5 6
6 7