OR-Tools v9.12: Use fixed intervals#175
Conversation
|
@dop-amin, are you planning to investigate the performance impact of this, or what are we going to do with this PR? |
|
I think using this feature of "newer" ortools versions is the right thing to do as it gives more information to the solver about the structure of the problem. I would think there is almost no way this could hurt the performance. I was waiting to investigate further until we've settled on a version of ortools back then. I don't have the time to follow up on the testing anymore. Maybe it's enough to rebase this and roughly check CI times before merging. |
Sounds like a good plan to me. Can you rebase it, please? |
|
I figured this feature seems to be new since 9.10(?) so merging now with 9.7 is incompatible. |
Agreed - we should not have separete paths for different ortools versions. |
3e1a852 to
1b19589
Compare
* According to a comment by lperron, this is supposed to improve efficiency of solving the constraint problem
Signed-off-by: Matthias J. Kannwischer <matthias@zerorisc.com>
1b19589 to
1f2eb54
Compare
This PR is based on #172.
According to a comment by Laurent Perron, this is supposed to improve efficiency of solving the constraint problem.
No thorough investigation about efficiency gains in our case has been done.