| Mon, 23 Aug 2010 23:00:57 +0200 | blanchet | cache previous iteration's weights, and keep track of what's dirty and what's clean; | changeset | files |
| Mon, 23 Aug 2010 21:55:52 +0200 | blanchet | modified relevance filter | changeset | files |
| Mon, 23 Aug 2010 21:11:30 +0200 | blanchet | adjust fudge factors in the light of experiments | changeset | files |
| Mon, 23 Aug 2010 18:53:11 +0200 | blanchet | invert semantics of "relevance_convergence", to make it more intuitive | changeset | files |
| Mon, 23 Aug 2010 18:39:12 +0200 | blanchet | if no facts were selected on first iteration, try again with a lower threshold | changeset | files |
| Mon, 23 Aug 2010 18:25:49 +0200 | blanchet | weed out junk in relevance filter | changeset | files |
| Mon, 23 Aug 2010 18:04:31 +0200 | blanchet | no need for "eq" facts -- good old "=" is good enough for us | changeset | files |