Re: pR and training mode

Bill Yerazunis wrote:
> Trever:
>
> one caution: it seems that (for reasons not yet fully understood)
> the current Hyperspace classifier is BROKEN.  This is established
> on 64-bit machines for sure.  
>
> You need to go back to a version that is "pre-VT", rather than the
> newest, latest, and greatest version.
>   
I am using the version provided by Fedora. crm114-0-1.7.20070810 is that 
pre VT or post VT?

I was doing tests this morning and it doesn't seem to be broken (on 
i386). Can you provide me with more details?
> But yes, your classifier statement is correct; the stats variable will
> get a nice little report as to the best matches.
>
>    - Bill Yerazunis
>
>   
Thank you, Bill. I greatly appreciate your help. This is great.

BTW, I have my html processor mostly working. I have had difficulty (as 
I stink with regex) figuring out how to remove <script xxxx /> and 
<script xxx>xxxx</script> tags with all between <script> and </script>. 
I also don't understand why I need isolate all over or I get duplicated 
entries and/or letters.

Thank you,

Trever
>    Date: Tue, 27 May 2008 08:13:29 -0600
>    From: "Trever L. Adams" <[email protected]>
>
>
>    Hello everyone,
>
>    I am having a difficult time finding answers to how to set the pR and=20
>    training mode. I am basing my attempt on some super script I found (the=20
>    only answer my google searches gave me).
>
>        learn <hyperspace unique> ssttt .5 (:*:_arg2:)
>
>    I am trying to do the best mode (ssttt, or is that ssttr?) and based on=20
>    what I am reading I need to use a pR of .25 to .5.  Am I doing this=20
>    correctly?
>
>    Also, is this the correct way to use it? And to classify?
>
>        classify <hyperspace unique > (A B C)  (:stats:)
>
>    Thank you,
>    Trever Adams
>
>    P.S. I have had some problems with my display for a few days, I will be=20
>    answering other help I have received in the next 36 hours.
>
>

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/

_______________________________________________
Crm114-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/crm114-general

{
    #  Remove all comments
    {
	match /<!--([^-]|-[^-]|--[^>])*-->/ (:whole_comment:)
	alter (:whole_comment:) //
	liaf
    }
    #  Turn all of the content of META tags into inline text
    {
	match <nocase> /<meta ([^>]*)>/ (:whole_meta_tag: :meta_txt:)
	{
		match <nocase> /http-equiv\=([^>]*)>/ (:meta_txt: :trash:)
		alter (:meta_txt:) //
	}
	{
		match <nocase> /content\=([^>]*)>/ (:meta_txt: :good:)
		alter (:meta_txt:) /:*:good:/
	}
	alter (:whole_meta_tag:) /:*:meta_txt:/
	liaf
    }
    #  Keep only the alt and title sections of an IMG tag
    {
	isolate (:quoted_title_txt: :unquoted_title_txt: :quoted_alt_txt: :unquoted_alt_txt:)
	alter (:quoted_title_txt:) //
	alter (:unquoted_title_txt:) //
	alter (:quoted_alt_txt:) //
	alter (:unquoted_alt_txt:) //
	match <nocase>  /<img ([^>]*)>/ (:whole_img_tag:)
	{
	    match <nocase> [:whole_img_tag:] /title="([^"]*)"/ (:trash: :quoted_title_txt:)
	}
	{
	    match <nocase> [:whole_img_tag:] /title=([^"][[:graph:]]*)/ (:trash: :unquoted_title_txt:) 
	}
	{
	    match <nocase> [:whole_img_tag:] /alt="([^"]*)"/ (:trash: :quoted_alt_txt:)
	}
	{
	    match <nocase> [:whole_img_tag:] /alt=([^"][[:graph:]]*)/ (:trash: :unquoted_alt_txt:)
	}
  	alter (:whole_img_tag:) / :*:quoted_alt_txt: :*:unquoted_alt_txt: :*:quoted_title_txt: :*:unquoted_title_txt: /
	liaf
    }
    #   Get rid of every other tag, without mercy!
    {
	match <nocase> /<[^>]*>/ (:some_tag:)
	alter (:some_tag:) //
	liaf
    }
    accept
}

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkg8GkIACgkQTmpiivh5HXNoGQCgmKbt1EUHk9abhA3FaOtYxh1j
EUMAnjvd7TP3JO9pT9TThJNZUcEu/Hua
=nqTE
-----END PGP SIGNATURE-----