gun tuning tangent

Have you tried using k=1? How does it compare then with something like regular kNN-PIF in terms of speed and hitrate?

Sorry, but i forgot details, everything that i remember i already wrote. Tomorrow i can publish that code, but i have no time in nearest future to liven up it

I guess K=1 would make ST-PIF have the same weaknesses as neural network based Pattern matching (non-statistical).

If a bot dodges 30% of the time going straight then turning to the right and 70% of the time going straight all the way, Neural Targeting averages both patterns and shoots slightly to the right, missing both patterns. In other words, it is awful against Walls.

Increasing K is what makes the gun choose the "straight all the way" pattern alone and achieve 70% hit rate.

Having a classifier which differentiates both patterns solves the problem (distance to corners against Walls).

If not, making many k-NN searches with k=1 gives you 30% of the ticks from one pattern and 70% of the ticks from the other pattern, and the resulting pattern is a 30%/70% mix of the two. It assumes a 3rd unseen pattern can be predicted from 2 previous patterns.

Should perform well against orbital movement, where turn rate changes gradually with distance and a 3rd unseen turn rate (or a zig-zag) can be predicted from 2 others. But perform bad against pattern movement, where patterns don't change gradually with classifiers.

I also thought on this problem and find out a possible solution: keep similar amount of data with different classes.