Difference between revisions of "Thread:Talk:DeBroglie/?????/reply (28)"
Jump to navigation
Jump to search
(Reply to ?????) |
|||
| Line 1: | Line 1: | ||
I´m using [[Wikipedia:Mahalanobis distance|normalized Euclidean distance]] (weight = 1/stddev) with [[Wikipedia:Standard deviation#Rapid calculation methods|fast sampled deviation]], which do change over time ([[Dynamic Clustering|k-NN]] for lazy people who don´t fine tune their bots). | I´m using [[Wikipedia:Mahalanobis distance|normalized Euclidean distance]] (weight = 1/stddev) with [[Wikipedia:Standard deviation#Rapid calculation methods|fast sampled deviation]], which do change over time ([[Dynamic Clustering|k-NN]] for lazy people who don´t fine tune their bots). | ||
| − | I had the impression the changing weights were not being accounted for. My [[Combat|bot´s]] | + | I had the impression the changing weights were not being accounted for. My [[Combat|bot´s]] scoring dramatically decreased with the [[Kd-tree|kd-tree]]. I suspect it is the cached coordinate limits inside each node. Or maybe I screwed up something while trying to use the library. |
Latest revision as of 23:04, 16 May 2012
I´m using normalized Euclidean distance (weight = 1/stddev) with fast sampled deviation, which do change over time (k-NN for lazy people who don´t fine tune their bots).
I had the impression the changing weights were not being accounted for. My bot´s scoring dramatically decreased with the kd-tree. I suspect it is the cached coordinate limits inside each node. Or maybe I screwed up something while trying to use the library.
