Age | Commit message (Collapse) | Author | |
---|---|---|---|
2018-08-12 | Fixed logic error on when nothing should be avoided | Justin Worthe | |
2018-08-12 | Eliminated assuming opponents might do nothing in random moves | Justin Worthe | |
2018-08-12 | Reduced more duplication and removed TODOs | Justin Worthe | |
2018-08-12 | Removed unused import | Justin Worthe | |
2018-08-12 | Folded duplicate code for player and opponent | Justin Worthe | |
2018-08-12 | Collapsed player info into the rest of the 'buildings' object | Justin Worthe | |
2018-08-12 | Bumped energy cutoffs right up to allow iron curtain | Justin Worthe | |
2018-08-12 | Test cases that do the iron curtain | Justin Worthe | |
2018-08-12 | Allowed monte carlo search to use iron curtains | Justin Worthe | |
2018-08-11 | Clippy-suggested edits | Justin Worthe | |
2018-08-09 | Rearranged random move selection to avoid expensive division | Justin Worthe | |
2018-08-09 | Removed unnecessary Player field | Justin Worthe | |
2018-08-09 | Removed dynamic settings | Justin Worthe | |
It worked really well for round 2 to set constants | |||
2018-08-09 | Added more TODO ideas | Justin Worthe | |
2018-07-24 | Fixed min and benchmark logging in discarding search | Justin Worthe | |
2018-07-24 | Modified pruning to be friendlier to calculating benchmarks | Justin Worthe | |
2018-07-23 | Added option for discarding poor performing moves early | Justin Worthe | |
2018-07-02 | Initial drop in replacement implementation | Justin Worthe | |
It's faster than the other one! Doesn't tesla yet, but still! Yay! | |||
2018-07-01 | Set up (failing) property test for working of new game state | Justin Worthe | |
Unfortunately, for this test to work, I still need to implement the function that reads the game state and goes from one to the other. | |||
2018-06-30 | Initial stab at putting game engine behind a trait | Justin Worthe | |
2018-06-30 | Removed deconstruction from the list of commands considered | Justin Worthe | |
I don't think deconstruction is a necessary part of a viable strategy for now. | |||
2018-06-25 | Added rule on maximum two tesla towers | Justin Worthe | |
2018-06-25 | Updated test cases and made engine work correctly according to tests | Justin Worthe | |
I don't think a tesla appeared in this match. I need to contrive a bot to build one. | |||
2018-06-25 | Compilation, allowing new moves to be chosen, and missile move order | Justin Worthe | |
2018-06-09 | Added pruning of buying energy buildings behind a feature flag | Justin Worthe | |
2018-06-02 | Moved replay-based test to have convenience import from game engine replay | Justin Worthe | |
2018-05-31 | Clippy suggested changes | Justin Worthe | |
2018-05-19 | Inlined unnecessary map | Justin Worthe | |
2018-05-19 | Removed dependency on OS randomness for seeding | Justin Worthe | |
I'm convinced that the cloud servers they're running on will be out of entropy, especially towards the end of the tournament. Rather have a less random initial seed than take that blocking performance hit. | |||
2018-05-16 | Moved away from special benchmarking suite | Justin Worthe | |
Just using normal monte carlo. More iterations -> better. | |||
2018-05-15 | Rayon for threading | Justin Worthe | |
2018-05-14 | Added running total of unoccupied cells | Justin Worthe | |
2018-05-14 | Changed invalid move checking to be a debug assertion | Justin Worthe | |
2018-05-14 | Reduced number of needless allocations to improve perf | Justin Worthe | |
Current iterations: 26486 in 10 seconds | |||
2018-05-14 | Added profiling target with perf | Justin Worthe | |
2018-05-12 | Limited bot to run within the 2 second window | Justin Worthe | |
2018-05-12 | Debugged and fixed the errors that had the monte carlo not working | Justin Worthe | |
Monte carlo now beats sample bot, if given plenty of time. I still need to put the max time tracking and enforcement in. | |||
2018-05-12 | Initial stab at monte carlo implementation | Justin Worthe | |
Doesn't seem to be working quite right... just sits there accumulating energy. |