Posted by Jared Anderson on July 28, 1999 at 13:11:30:
For the in-class example of Floyd's algorithm, we had a nice neat complete graph, where all nodes were directly adjacent to all other nodes. This really simplified the algorithm.
However, for the HW, there are certain paths that do NOT start out "direct".
When we're developing our table, where all possible edges, does this mean, for example, for node 1 we'd be listing only 1-2, 1-4, and 1-6? If this is the case, then after the table is complete, we haven't yet addressed, for example, 1-3 or 1-5, or others such as 2-5, 2-6, etc.
So, what is it THEN a matter of? How do we take what we've gotten from the table and apply it to those "other paths" ... I imagine there's a better way to do it than just "trying all combinations"?
Or, are we to somehow include, instead of "all possible EDGES", "all possible PATHS" in our ORIGINAL table, where 1-4 and 1-5 would be included and represent the paths that start at 1 and end at 4/5 ... At first, until a relevant node was processed, would this be "INFINITY" as with Dijkstra's?
... Am I making ANY sense?? lol
--Thanks