On 25 June 2012 04:13, Philippe Michel
<address@hidden> wrote:
On Sun, 24 Jun 2012, Joseph Heled wrote:
On 24 June 2012 13:38, Joseph Heled <address@hidden> wrote:
Hi,
It is heart warming to see that the race net can be improved still :)
The benchmark rates the new crashed net as stronger, which is great too,
but the new contact net is rated weaker in moves (0.0104453134853 vs
0.0104996763558), stronger in cube actions.
I take this back. After updating the contact benchmark to include all
missing positions, the new net comes out ahead. The crashed benchmark need
to be updated as well.
For the crashed benchmark I made what looked like an easy try to alleviate this by rounding up the number of rolled-out alternatives to the next multiple of 5 but I don't know if it made much of a difference.
I did that because it looked right on general principles but I totally missed the fact that it may matter to explain small progresses, or the lack of them :-(.
I suppose there is an easy way to find and add them from a verbose benchmark output since you apparently did this quickly although there must be hundreds of them.
Yes, perr has a --log option, whose output can be fed to sa
-Joseph