bogotuning

[Allison, Tom]

Greetings.

I'm trying to do some tuning, but there are two things that prevent me from doing this.

First, I don't have quite enough messages but I'm working on that.  Given time, I'm sure I'll have enough.

But the other is the message I get from my bogotune output:

bogotune -n ~/ham/ -s ~/spam/
The wordlist contains 2016 non-spam and 1268 spam messages.
Bogotune must be run with at least 2000 of each.
wordlist's ham to spam ratio is 1.6 to 1.0
The messages sets contain 0 non-spam and 0 spam.  Bogotune requires at least 500 non-spam and 500 spam messages to run.

Each directory, ham/ & spam/, are full of email (2016 & 1268 respectively) but I don't understand the comment that the messages sets contain 0 non-spam and 0 spam.

Not sure if it matters, but each message is stored as raw text, not really following any specific mail delivery format.  thought I would mention this because it's atypical.