# TestConditionalTraining: By default, dspam will retrain certain errors
# until the condition is no longer met. This usually accelerates
learning.
# Some people argue that this can increase the risk of errors, however.
#
TestConditionalTraining off
If you have test conditional on, after training a message once dspam
checks to make sure that a classify will give the correct answer. If
it does not then it trains it again (for spam it decrements spam
tokens & increments innocent tokens). Each training loop after the
first one gets you a corpus fed hit.
gab
On Apr 4, 2008, at 3:17 PM, john espiro wrote:
In the webui, under From Corpus, I am seeing a number of 5232 for
SPAM messages fed, and 0 under good messages.
I never fed any messages from a corpus so I wonder what this field
means and where it would get this number. Any ideas?
Thanks,
-- John
You rock. That's why Blockbuster's offering you one month of
Blockbuster Total Access, No Cost.
