From gevans58 at gmail.com  Mon Jan 14 03:10:37 2013
From: gevans58 at gmail.com (George Evans)
Date: Sun, 13 Jan 2013 21:10:37 -0500
Subject: [Spambayes] Outlook 2013
Message-ID: <000001cdf1fc$597bf180$0c73d480$@gmail.com>

Is SpamBayes now working with the latest Outlook - version 2013? I didn't
see it referenced on your website.

 

Thanks,

 

George Evans

 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/spambayes/attachments/20130113/629e2732/attachment.html>

From thruska at cubiclesoft.com  Mon Jan 14 03:48:08 2013
From: thruska at cubiclesoft.com (Thomas Hruska)
Date: Sun, 13 Jan 2013 19:48:08 -0700
Subject: [Spambayes] Outlook 2013
In-Reply-To: <000001cdf1fc$597bf180$0c73d480$@gmail.com>
References: <000001cdf1fc$597bf180$0c73d480$@gmail.com>
Message-ID: <50F371E8.5040107@cubiclesoft.com>

On 1/13/2013 7:10 PM, George Evans wrote:
> Is SpamBayes now working with the latest Outlook - version 2013? I didn't
> see it referenced on your website.
>
> Thanks,
>
> George Evans

A good rule of thumb to follow is that if a product supports the 
previous version of an OS or product, it will generally work under the 
latest version of that same product.  Microsoft tends to not rock the 
boat with Windows and Office.

That said, the Spambayes Outlook plugin, from what I've read, has issues 
with 64-bit Windows/Outlook.  The developers generally don't care for 
Windows users (*NIX heads), so their interest in supporting the Outlook 
plugin is, well, pretty much non-existent.  And, if they could have 
their way, they'd like to not support Windows users either despite 
Windows still dominating desktop market share.

This mailing list is quiet.  That could be related to the above or it 
could be related to external factors.  It also doesn't help that there 
has been no public-facing work done on the product since 2011 - even if 
it is to update copyright years in the source, that's still something, 
albeit lazy.  I usually take that as a sign of a dead project or headed 
quickly toward dead project status.

-- 
Thomas Hruska
CubicleSoft President

I've got great, time saving software that you might find useful.

http://cubiclesoft.com/

From railmeat at gmail.com  Mon Jan 14 04:34:35 2013
From: railmeat at gmail.com (Matt Johnson)
Date: Sun, 13 Jan 2013 19:34:35 -0800
Subject: [Spambayes] Outlook 2013
In-Reply-To: <000001cdf1fc$597bf180$0c73d480$@gmail.com>
References: <000001cdf1fc$597bf180$0c73d480$@gmail.com>
Message-ID: <CAPou7eD0AAb+EKyDnGbnKf9LS9PArPqy=O8gD-TqW2=GJe4qLA@mail.gmail.com>

George, probably the only way to find out is to install it and see
what happens. If you do that please share what you learn.

As Thomas said the project seems to be rather quit.

--
Matt Johnson


On Sun, Jan 13, 2013 at 6:10 PM, George Evans <gevans58 at gmail.com> wrote:
> Is SpamBayes now working with the latest Outlook - version 2013? I didn?t
> see it referenced on your website.
>
>
>
> Thanks,
>
>
>
> George Evans
>
>
>
>
>
>
> _______________________________________________
> SpamBayes at python.org
> http://mail.python.org/mailman/listinfo/spambayes
> Info/Unsubscribe: http://mail.python.org/mailman/listinfo/spambayes
> Check the FAQ before asking: http://spambayes.sf.net/faq.html

From skip at pobox.com  Mon Jan 14 05:13:00 2013
From: skip at pobox.com (Skip Montanaro)
Date: Sun, 13 Jan 2013 22:13:00 -0600
Subject: [Spambayes] Outlook 2013
In-Reply-To: <50F371E8.5040107@cubiclesoft.com>
References: <000001cdf1fc$597bf180$0c73d480$@gmail.com>
	<50F371E8.5040107@cubiclesoft.com>
Message-ID: <CANc-5UxQpjseC8xLwVMRa6ydoB-+C2U7jFzFG28QYC6tE+boUg@mail.gmail.com>

> The developers generally don't care for
> Windows users (*NIX heads), so their interest in supporting the Outlook
> plugin is, well, pretty much non-existent.  And, if they could have their
> way, they'd like to not support Windows users either despite Windows still
> dominating desktop market share.

I think that characterization is more than a bit unfair.  For a long
while, several people actively worked to support Windows.  That
Windows support has dwindled is more an indication that they have no
time left to support SpamBayes (regardless of platform) than that they
are Unix snobs.

Perhaps you have forgotten that SpamBayes is open source software.
You (or anyone else) are more than welcome to grab the source and do
what needs to be done to get SpamBayes working again with more recent
versions of Windows.  I have put out the plea for developers various
places more than once and have always been met with silence.

As one of the lesser SpamBayes developers, I actually find I no longer
need SpamBayes.  Over time, the various organizations which provide me
with my daily diet of email (pobox.com and Gmail) have gotten better
and better at filtering out spam.  Before I actually stopped using it,
those providers did such a good job that it was pretty rare that the
software was even exercised.  I suspect some of the other SpamBayes
developers may also be former SpamBayes users.  Time marches on.

Skip Montanaro

From thruska at cubiclesoft.com  Mon Jan 14 12:15:49 2013
From: thruska at cubiclesoft.com (Thomas Hruska)
Date: Mon, 14 Jan 2013 04:15:49 -0700
Subject: [Spambayes] Outlook 2013
In-Reply-To: <CANc-5UxQpjseC8xLwVMRa6ydoB-+C2U7jFzFG28QYC6tE+boUg@mail.gmail.com>
References: <000001cdf1fc$597bf180$0c73d480$@gmail.com>
	<50F371E8.5040107@cubiclesoft.com>
	<CANc-5UxQpjseC8xLwVMRa6ydoB-+C2U7jFzFG28QYC6tE+boUg@mail.gmail.com>
Message-ID: <50F3E8E5.20205@cubiclesoft.com>

On 1/13/2013 9:13 PM, Skip Montanaro wrote:
> Perhaps you have forgotten that SpamBayes is open source software.
> You (or anyone else) are more than welcome to grab the source and do
> what needs to be done to get SpamBayes working again with more recent
> versions of Windows.  I have put out the plea for developers various
> places more than once and have always been met with silence.

This is where you officially move the project(s) to GitHub and pay 
attention to pull requests.  You're more likely to get people 
contributing on GitHub than SourceForge.  Forking, modifying, and 
submitting pull requests is just as easy as merging and accepting the 
pull request into the main branch.  If you don't want to do any 
development (the hard part anyway), the key is to stay on top of pull 
requests and don't let them sit around in the queue for more than a 
couple weeks.  The work on your end becomes rather minimal - taking more 
of a hands-off managerial role.

Automating builds is also important so that someone can grab the latest 
binaries and run them.  Once the automated system works, it is also 
fairly hands-off.


> As one of the lesser SpamBayes developers, I actually find I no longer
> need SpamBayes.  Over time, the various organizations which provide me
> with my daily diet of email (pobox.com and Gmail) have gotten better
> and better at filtering out spam.  Before I actually stopped using it,
> those providers did such a good job that it was pretty rare that the
> software was even exercised.  I suspect some of the other SpamBayes
> developers may also be former SpamBayes users.  Time marches on.

I suspected this was the case.  A move to GitHub would at least afford 
new life into the product.  There's always room for improvement. 
Spambayes is still useful as a tool in corporate environments where most 
companies have their own mail servers and half of the anti-spam 
appliances are terrible at dealing with spam (legit mail gets hung up 
more often than not and spam leaks like a sieve).  For example, Postini 
is owned by Google, but it does a rather terrible job at filtering spam. 
  Google has plans to shut it down though.  At that point, companies 
might come running to solutions like this if they don't like the 
results.  Positioning this project for that possible eventuality would 
be a smart move.

The key to developing software is to use it daily.  You might want to 
consider dropping pobox.com and Gmail if their anti-spam measures are so 
good that you no longer feel the need to develop this product. 
Enterprises still need good server and client-side anti-spam tools and 
that likely won't change any time soon.  To that end, this product 
should shift focus from meeting individual needs to the needs of the 
enterprise.

-- 
Thomas Hruska
CubicleSoft President

I've got great, time saving software that you might find useful.

http://cubiclesoft.com/

From jonas at rcpt.at  Fri Jan 18 09:03:23 2013
From: jonas at rcpt.at (Jonas)
Date: Fri, 18 Jan 2013 09:03:23 +0100
Subject: [Spambayes] Training on data using timcv.py
Message-ID: <20130118090323.aox5xvxo9sg8co80@webmail.ira.uni-karlsruhe.de>

Hi,
for my research project, I need to train spambayes using one of the  
training tools.
I have two mboxes (ham/spam). I splitted them up using splitndirs.py,  
which results in
Data/Ham/Set.. and Data/Spam/Set.. .
When I execute "timcv.py -n 5" I receive the following errors:

....
Traceback (most recent call last):
   File "C:\spambayes-src\spambayes\timcv.py", line
  175, in <module>
     main()
   File "C:\spambayes-src\spambayes\timcv.py", line
  172, in main
     drive(nsets)
   File "C:\spambayes-src\spambayes\timcv.py", line
  82, in drive
     d = TestDriver.Driver()
   File  
"C:\Python\64Bit\lib\site-packages\spambayes-1.1b1-py2.7.egg\spambayes\Te
stDriver.py", line 135, in __init__
     self.new_classifier()
   File  
"C:\Python\64Bit\lib\site-packages\spambayes-1.1b1-py2.7.egg\spambayes\Te
stDriver.py", line 141, in new_classifier
     self.set_classifier(classifier.Bayes())
   File  
"C:\Python\64Bit\lib\site-packages\spambayes-1.1b1-py2.7.egg\spambayes\Te
stDriver.py", line 146, in set_classifier
     self.tester = Tester.Test(classifier)
TypeError: __init__() takes exactly 1 argument (2 given)

I could not figure out the error by checking the source code of timcv.py and
Testdriver.py.
I receive the same error, when I execute timtest.py

Any ideas how to solve the problem? I downloaded spambayes from the  
following source (
https://spambayes.svn.sourceforge.net/svnroot/spambayes/trunk/spambayes/ ) and  
installed
it with "setup.py install".

Jonas

From s_prior at ira.uka.de  Fri Jan 18 09:01:36 2013
From: s_prior at ira.uka.de (s_prior at ira.uka.de)
Date: Fri, 18 Jan 2013 09:01:36 +0100
Subject: [Spambayes] Training on data using timcv.py
Message-ID: <20130118090136.c6dekg7m4g4og80g@webmail.ira.uni-karlsruhe.de>

Hi,
for my research project, I need to train spambayes using one of the  
training tools.
I have two mboxes (ham/spam). I splitted them up using splitndirs.py,  
which results in Data/Ham/Set.. and Data/Spam/Set.. .
When I execute "timcv.py -n 5" I receive the following errors:

....
Traceback (most recent call last):
   File "C:\spambayes-src\spambayes\timcv.py", line
  175, in <module>
     main()
   File "C:\spambayes-src\spambayes\timcv.py", line
  172, in main
     drive(nsets)
   File "C:\spambayes-src\spambayes\timcv.py", line
  82, in drive
     d = TestDriver.Driver()
   File  
"C:\Python\64Bit\lib\site-packages\spambayes-1.1b1-py2.7.egg\spambayes\Te
stDriver.py", line 135, in __init__
     self.new_classifier()
   File  
"C:\Python\64Bit\lib\site-packages\spambayes-1.1b1-py2.7.egg\spambayes\Te
stDriver.py", line 141, in new_classifier
     self.set_classifier(classifier.Bayes())
   File  
"C:\Python\64Bit\lib\site-packages\spambayes-1.1b1-py2.7.egg\spambayes\Te
stDriver.py", line 146, in set_classifier
     self.tester = Tester.Test(classifier)
TypeError: __init__() takes exactly 1 argument (2 given)

I could not figure out the error by checking the source code of  
timcv.py and Testdriver.py.
I receive the same error, when I execute timtest.py

Any ideas how to solve the problem? I downloaded spambayes from the  
following source (  
https://spambayes.svn.sourceforge.net/svnroot/spambayes/trunk/spambayes/ ) and  
installed it with "setup.py install".

Jonas



From jonas at rcpt.at  Sat Jan 19 13:02:56 2013
From: jonas at rcpt.at (Jonas)
Date: Sat, 19 Jan 2013 13:02:56 +0100
Subject: [Spambayes] Training on data using timcv.py
In-Reply-To: <20130118090323.aox5xvxo9sg8co80@webmail.ira.uni-karlsruhe.de>
References: <20130118090323.aox5xvxo9sg8co80@webmail.ira.uni-karlsruhe.de>
Message-ID: <20130119130256.nwg6fpo6ooscwskk@webmail.ira.uni-karlsruhe.de>

Hi,
some more information about my installation (installed it both on  
linux and windows 7 64bit).

1. Downloaded and unzipped  
http://sourceforge.net/projects/spambayes/files/spambayes/1.1a6/spambayes-1.1a6.zip/download

2. Installed it with python setup.py install

3. started server in scripts/sb_server.py

4. Splitted mboxes with splitndirs.py

5. python timcv.py -n 5

I also installed the  1.14  of Spambayes, but I still receive the same error.


Any  suggestions? :)

Jonas
Zitat von Jonas <jonas at rcpt.at>:

> Hi,
> for my research project, I need to train spambayes using one of the
> training tools.
> I have two mboxes (ham/spam). I splitted them up using splitndirs.py,
> which results in
> Data/Ham/Set.. and Data/Spam/Set.. .
> When I execute "timcv.py -n 5" I receive the following errors:
>
> ....
> Traceback (most recent call last):
>   File "C:\spambayes-src\spambayes\timcv.py", line
>  175, in <module>
>     main()
>   File "C:\spambayes-src\spambayes\timcv.py", line
>  172, in main
>     drive(nsets)
>   File "C:\spambayes-src\spambayes\timcv.py", line
>  82, in drive
>     d = TestDriver.Driver()
>   File
> "C:\Python\64Bit\lib\site-packages\spambayes-1.1b1-py2.7.egg\spambayes\Te
> stDriver.py", line 135, in __init__
>     self.new_classifier()
>   File
> "C:\Python\64Bit\lib\site-packages\spambayes-1.1b1-py2.7.egg\spambayes\Te
> stDriver.py", line 141, in new_classifier
>     self.set_classifier(classifier.Bayes())
>   File
> "C:\Python\64Bit\lib\site-packages\spambayes-1.1b1-py2.7.egg\spambayes\Te
> stDriver.py", line 146, in set_classifier
>     self.tester = Tester.Test(classifier)
> TypeError: __init__() takes exactly 1 argument (2 given)
>
> I could not figure out the error by checking the source code of timcv.py and
> Testdriver.py.
> I receive the same error, when I execute timtest.py
>
> Any ideas how to solve the problem? I downloaded spambayes from the
> following source (
> https://spambayes.svn.sourceforge.net/svnroot/spambayes/trunk/spambayes/ )   
> and
> installed
> it with "setup.py install".
>
> Jonas
> _______________________________________________
> SpamBayes at python.org
> http://mail.python.org/mailman/listinfo/spambayes
> Info/Unsubscribe: http://mail.python.org/mailman/listinfo/spambayes
> Check the FAQ before asking: http://spambayes.sf.net/faq.html



From jonas at rcpt.at  Mon Jan 21 12:12:00 2013
From: jonas at rcpt.at (Jonas)
Date: Mon, 21 Jan 2013 12:12:00 +0100
Subject: [Spambayes] Training on data using timcv.py
In-Reply-To: <20130118090323.aox5xvxo9sg8co80@webmail.ira.uni-karlsruhe.de>
References: <20130118090323.aox5xvxo9sg8co80@webmail.ira.uni-karlsruhe.de>
Message-ID: <20130121121200.ieqkkgls844gocwo@webmail.ira.uni-karlsruhe.de>

Hi,
I think, there is a bug in TestDriver.py:
I modified line 146 and added another line:
Previosly it was:
        146     self.tester = Tester.Test(classifier)

Now:
       146  self.tester = Tester.Test()
       147  self.tester.set_classifier(classifier);


timcv.py can be now executed.


Jonas

Zitat von Jonas <jonas at rcpt.at>:

> Hi,
> for my research project, I need to train spambayes using one of the
> training tools.
> I have two mboxes (ham/spam). I splitted them up using splitndirs.py,
> which results in
> Data/Ham/Set.. and Data/Spam/Set.. .
> When I execute "timcv.py -n 5" I receive the following errors:
>
> ....
> Traceback (most recent call last):
>   File "C:\spambayes-src\spambayes\timcv.py", line
>  175, in <module>
>     main()
>   File "C:\spambayes-src\spambayes\timcv.py", line
>  172, in main
>     drive(nsets)
>   File "C:\spambayes-src\spambayes\timcv.py", line
>  82, in drive
>     d = TestDriver.Driver()
>   File
> "C:\Python\64Bit\lib\site-packages\spambayes-1.1b1-py2.7.egg\spambayes\Te
> stDriver.py", line 135, in __init__
>     self.new_classifier()
>   File
> "C:\Python\64Bit\lib\site-packages\spambayes-1.1b1-py2.7.egg\spambayes\Te
> stDriver.py", line 141, in new_classifier
>     self.set_classifier(classifier.Bayes())
>   File
> "C:\Python\64Bit\lib\site-packages\spambayes-1.1b1-py2.7.egg\spambayes\Te
> stDriver.py", line 146, in set_classifier
>     self.tester = Tester.Test(classifier)
> TypeError: __init__() takes exactly 1 argument (2 given)
>
> I could not figure out the error by checking the source code of timcv.py and
> Testdriver.py.
> I receive the same error, when I execute timtest.py
>
> Any ideas how to solve the problem? I downloaded spambayes from the
> following source (
> https://spambayes.svn.sourceforge.net/svnroot/spambayes/trunk/spambayes/ )   
> and
> installed
> it with "setup.py install".
>
> Jonas
> _______________________________________________
> SpamBayes at python.org
> http://mail.python.org/mailman/listinfo/spambayes
> Info/Unsubscribe: http://mail.python.org/mailman/listinfo/spambayes
> Check the FAQ before asking: http://spambayes.sf.net/faq.html



From whille at 163.com  Fri Jan 25 04:34:44 2013
From: whille at 163.com (whille)
Date: Fri, 25 Jan 2013 11:34:44 +0800
Subject: [Spambayes] question about bayes classifier application
Message-ID: <5101FD54.5010909@163.com>

Hi all,
      i've just signed in the email group. please tell me if any mistakes.
      i'm asking suggestion for my testing result.

      i'm using spambayes src codes, to apply in person classifying: 
personal or organization.
      basically use Tester and Bayes codes:

from spambayes.classifier import Bayes
from spambayes.Options import options
from spambayes.Tester import Test, _Example

tokenizing sample, add 'user:' to username field:
         username: user:...
         description: ...
         vip: True/False

the result i got now is appended,  400~4k samples used to train and predict.
method: use 90% to train, rest 10% to test. average 10 times to get result.

5-5-img3.png:
options["Categorization", "ham_cutoff"] = 0.5
options["Categorization", "spam_cutoff"] = 0.5
false_positive_rate still climbing at 4k samples.

4-8.5-....png:
options["Categorization", "ham_cutoff"] = 0.4
options["Categorization", "spam_cutoff"] = 0.85
unsure_rate still climbing. false_positive_rate climbing slowly. If add 
both, result in similar as 0.5-.0.5 cutoff

Is this normal in spambayes app? Any wrong  in my usage? thanks a lot.


-------------- next part --------------
A non-text attachment was scrubbed...
Name: 5-5-img3.png
Type: image/png
Size: 36776 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/spambayes/attachments/20130125/76166896/attachment-0002.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 4-8.5-img-vtype-imgcomplex.png
Type: image/png
Size: 36984 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/spambayes/attachments/20130125/76166896/attachment-0003.png>