http://bugzilla.spamassassin.org/show_bug.cgi?id=3671





------- Additional Comments From [EMAIL PROTECTED]  2004-08-06 22:22 -------
Subject: Re:  Possible 30 - 40% speed increase for Bayes with SDBM_File

FYI, I tried an sa-learn on !12000 msgs, ham and spam, here are the
results:

Took about 11 mins

Learned from 2000 message(s) (2000 message(s) examined).
Learned from 2000 message(s) (2000 message(s) examined).
Learned from 2000 message(s) (2000 message(s) examined).
Learned from 2000 message(s) (2000 message(s) examined).
Learned from 1999 message(s) (2000 message(s) examined).
Learned from 1999 message(s) (2000 message(s) examined).
$ sa-learn --dump magic
0.000          0          3          0  non-token data: bayes db version
0.000          0       5998          0  non-token data: nspam
0.000          0       5997          0  non-token data: nham
0.000          0     541953          0  non-token data: ntokens
0.000          0 1077432565          0  non-token data: oldest atime
0.000          0 1087258934          0  non-token data: newest atime
0.000          0          0          0  non-token data: last journal sync atime
0.000          0          0          0  non-token data: last expiry atime
0.000          0          0          0  non-token data: last expire atime delta
0.000          0          0          0  non-token data: last expire reduction 
count

-rw-------    1 parker   users        4096 2004-08-06 23:53 bayes_seen.dir
-rw-------    1 parker   users     2096128 2004-08-06 23:53 bayes_seen.pag
-rw-------    1 parker   users        4096 2004-08-06 23:53 bayes_toks.dir
-rw-------    1 parker   users    16775168 2004-08-06 23:53 bayes_toks.pag

In comparison, here is the same data via DBM:

-rw-------    1 parker   users     1306624 2004-08-07 00:21 bayes_seen
-rw-------    1 parker   users    20930560 2004-08-07 00:21 bayes_toks

Michael





------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

Reply via email to