Author: Pete R.
Date: 16:38:01 07/25/00
Go up one level in this thread
On July 25, 2000 at 18:58:45, Pete R. wrote: >However, couldn't you simply use existing test suites for the same purpose with >just a little tweaking? For example suppose I am reasonably certain that a >given test suite position has the correct best move listed. It follows then >that if you generate all legal moves from that test position, the condition you >want is that the position resulting from the best move has the highest eval. In >other words your program will take the test position, generate one position for >each legal move from there and run the eval on each resulting position, and see >if in fact your eval weights the position resulting from the best move higher >than the rest. Small correction, of course it's not necessary to compute positions on the fly, it would be faster and simpler to determine the reasonable candidate moves for each test position beforehand, and then simply check that the eval weight is highest of those few candidates. The more "positional" the test positions are the better. Unless I have my head stuck somewhere, which is always a possibility.
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.