aboutsummaryrefslogtreecommitdiffstats
path: root/src/classifiers
Commit message (Collapse)AuthorAgeFilesLines
* * Fix awfull bug in classifying when first statfile has twice weight than ↵Vsevolod Stakhov2010-05-141-2/+2
| | | | | | second... * Fix undisclosed recipients detection
* * Bugfixes:Vsevolod Stakhov2010-04-201-1/+1
| | | | | | | | | | - handle '\' characters in lua strings correctly - fix lua initialization - avoid of using global lua state (global L) - fix listen sockets hash to allow multiply workers of same type but on different listen sockets - fix modules options inserting to allow multiply options of the same name - fix parsing of lua options - fix lua rules
* * Add option min_tokens to classifier that allows to skip too short messages ↵Vsevolod Stakhov2010-03-221-1/+34
| | | | | | from statistic check, format: min_tokens = "10"; (for 10 words minimum)
* * Fix bugs from previous commitcebka@lenovo-laptop2010-03-011-6/+3
|
* * Add weights command for getting weights of each message by each statfilecebka@lenovo-laptop2010-03-013-6/+80
| | | | | * Add ability to specify multiplier when learning * Add statistics about spam and ham messages
* * Forgotten call of normalizer functioncebka@lenovo-laptop2010-01-141-0/+3
|
* * Introduce new logging system:Vsevolod Stakhov2009-12-221-1/+1
| | | | | | | | | - independent and customizeable buffering - line buffering - errors handling support - custom (ip based) debug - append function name automaticaly (based on __FUNCTION__) - add some logic to logs system
* * Implement pre and post classify callbacks for checking specific statfiles ↵Vsevolod Stakhov2009-12-161-3/+21
| | | | | | | | | | for this task TODO: - add properties to get all parameters of input task - add properties to statfile object - add some normalization function for calling from classify process - document changes
* * Fix symbols cache (init lua filters before symbols cache initialization)Vsevolod Stakhov2009-12-141-3/+21
| | | | | | | * Remove LRU expiration logic from statfiles and replace it with random/lowerest value expiration logic: expire random block or block with lowerest value ! statfiles are incompatible again
* * Many major fixes to statfiles:Vsevolod Stakhov2009-12-032-18/+31
| | | | | | | | | | - fix bug with mmapping files: new addresses must NOT be allocated in shared memory by themselves - fix bug with winnow classifier that totally brokes it down - fix bug with too much grow of values * Use double precission values in statistics * Add statistics for statfiles * Add more informative data to output of LEARN command (weight of incoming message) * Add weight to output of classifier as well
* * Write revision and revision time to statfileVsevolod Stakhov2009-11-122-24/+4
| | | | * Make some improvements to API (trying to make it more clear)
* * Add binlog API implementationVsevolod Stakhov2009-11-061-1/+3
|
* * Add ability to change statfile size limit in config and allow reindexing ↵Vsevolod Stakhov2009-10-161-4/+4
| | | | of statfiles
* * Retab, no functional changesVsevolod Stakhov2009-10-022-53/+53
|
* * Fix race between learn and classifyVsevolod Stakhov2009-09-281-1/+5
|
* * Fix learningVsevolod Stakhov2009-09-251-3/+5
|
* * Remove assertVsevolod Stakhov2009-09-161-3/+5
| | | | | | * Fix build WITH_LUA * Fix calling of classifier * Fix autolearn
* * New system of classifiers interface and statfiles processingVsevolod Stakhov2009-09-143-67/+66
| | | | | | | * Fix sample config * Fix compile warnings * Fix building without lua support * Fix bugs with nrcpt header parsing and symbols cache loading (by Anton Nekhoroshikh)
* * Rework structure and API of statfiles functions to improve performance and ↵Vsevolod Stakhov2009-07-021-13/+11
| | | | | | avoid missusage of hash table * Correct url length calculation in urls command
* * Prepare to migrate to cmake (still need to write install target and ↵Vsevolod Stakhov2009-02-163-9/+2
| | | | | | | | working with XS implicitly) * Move all system includes to one file where we detect availability of all that includes * Fix license misprint * Fix some issues with perl initializing
* * Add BSD license textVsevolod Stakhov2009-02-162-0/+48
|
* * Avoid division by zeroVsevolod Stakhov2009-02-091-1/+8
|
* * Do another rework of filters/metrics and statfiles processingVsevolod Stakhov2009-01-293-17/+83
| | | | * Add 'probability' normalizing to winnow algorithm and calculate not pure weight but normalized probability
* * Rework statfiles result processingVsevolod Stakhov2009-01-223-1/+9
| | | | * Fix small bug in protocol implementation (\r -> \r\n)
* * Add simple implementation of classifiers abstraction and winnow classifierVsevolod Stakhov2008-12-043-0/+168
* Force statfile to work with float values