Search found 12 matches
- 28 Mar 2019, 11:24
- Forum: Infrastructure Development
- Topic: refactoring repos.springrts.com (rapid repositories)
- Replies: 22
- Views: 15744
Re: Experimenting with refactoring repos.springrts.com
Edit: ZK has its own downloader that failed due to this update.
- 02 Feb 2019, 21:42
- Forum: SPADS AutoHost
- Topic: TrueSkill 2 vs WHR
- Replies: 16
- Views: 14449
Re: TrueSkill 2 vs WHR
You should train + test it on at least a few hundred games and be able to show a better predictive accuracy that the existing system. Also make sure to check the significance level. With 1000 games there's still a std dev of +-1.6% around 50%. Differences in rating systems can often be smaller than...
- 01 Feb 2019, 22:43
- Forum: SPADS AutoHost
- Topic: TrueSkill 2 vs WHR
- Replies: 16
- Views: 14449
Re: TrueSkill 2 vs WHR
Also, ranking computations need to be reasonably fast so that it's possible to take into account smurfs detections and manual accounts split/join retro-actively, which can lead to recomputing all matches since 2012 on the fly. While WHR can't detect smurfs, it can be fun to confirm your predictions...
- 01 Feb 2019, 22:21
- Forum: SPADS AutoHost
- Topic: TrueSkill 2 vs WHR
- Replies: 16
- Views: 14449
Re: TrueSkill 2 vs WHR
This isn't (always) correct - for example the TS2 paper tests their model by comparing "expected win rates" to actual win rates. Since the model makes non-deterministic preditions of winners, the expected win rate is the average across the test data of predicted winning probabilities per ...
- 31 Jan 2019, 22:42
- Forum: SPADS AutoHost
- Topic: TrueSkill 2 vs WHR
- Replies: 16
- Views: 14449
Precision vs Accuracy
Convergence Firstly, there is how data points are used. Independent of what data is actually collected, both systems make different use of the data. Trueskill 1/2 always only considers the most recent result and most recent ratings associated with each player. This makes it very easy to both calcul...
- 09 Apr 2017, 15:26
- Forum: AI
- Topic: Connecting SkirmishAI to the current player
- Replies: 2
- Views: 5656
Re: Connecting SkirmishAI to the current player
/AIControl (unsynced) Creates a new instance of a Skirmish AI, to let it controll a specific team /AIKill (unsynced) Kills the Skirmish AI controlling a specified team /AIList (unsynced) Prints a list of all currently active Skirmish AIs /AIReload (unsynced) Reloads the Skirmish AI controlling a sp...
- 09 Apr 2017, 01:51
- Forum: AI
- Topic: Connecting SkirmishAI to the current player
- Replies: 2
- Views: 5656
Connecting SkirmishAI to the current player
Now that I've spent quite some time developing a SkirmishAI, I'd be interested in expanding its usefulness as an extension of ingame widgets. What would be required to allow AIs to take control of the team one is playing in? Could this be done without too much hassle within Spring? An alternative op...
- 09 Apr 2017, 01:36
- Forum: AI
- Topic: Profiling Java AI
- Replies: 4
- Views: 7641
Re: Profiling Java AI
You can simply use the NetBeans Profiler. It shows you the time spent in each method, memory usage per object and some other helpful stats. It helped me a lot in optimizing CSI.
- 16 Jul 2016, 00:33
- Forum: Spring Clans
- Topic: [GBC] Clan
- Replies: 6
- Views: 10098
Re: [GBC] Clan
seems some conspirators got drunk..
- 17 Dec 2015, 00:11
- Forum: AI
- Topic: [Java] Clear factory build queue [solved]
- Replies: 4
- Views: 3713
Re: [Java] Clear factory build queue
Thanks for the quick response. Although replacing 64 with 16(RMB) doesn't alter the queue :/
Edit: After using 255 (all possible modifier keys mashed at once) it worked :D
Edit: After using 255 (all possible modifier keys mashed at once) it worked :D
- 16 Dec 2015, 22:24
- Forum: AI
- Topic: [Java] Clear factory build queue [solved]
- Replies: 4
- Views: 3713
[Java] Clear factory build queue [solved]
Is there a simple way to clear the build queue of a factory? I have tried unit.stop() with no success. List<Float> cmds;//contains CommandIDs(=negative unit def id) of units in queue getUnit().executeCustomCommand(2, cmds, 64, frame + 50) doesn't seem to work either. 2 indicates CMD_REMOVE, 64 stand...
- 14 Aug 2014, 00:50
- Forum: AI
- Topic: enemy attack prediction
- Replies: 14
- Views: 3134
Re: enemy attack prediction
what im currently using is maybe not exactly what you mean by quad(tree?) map but should work similarly. the idea was to make the whole ai based on this map. ill just explain what zkmbai is using in case someone is interested or wants to help me improve it: the ai divides the map into 8x8 rectangles...