Michael Zarozinski

Show:

What's happening?

  • Followup: RE: Multiple Output Variables

    Matthew, The parsing of FCL is rather brain dead... the parser is not very robust, it was coded just to handle the specifics of FCL. The repeated points are due to that term being a 'S Curve'... those require 7 points. Any variance in what the parser expects (removing repeated points for example) can cause the parser to act up. It was my hope that as an open source project someone...

    2006-01-27 15:21:36 UTC in FFLL - Free Fuzzy Logic Library

  • Followup: RE: Multiple Output Variables

    Matthew, Sorry for the late reply. It does appear that FCL allows for more than one output variable, but FFLL was designed for one output var, there's nothing in theory that would prevent > 1 output var, (off the top of my head) you'd need to modify FuzzyModelBase to have an array of output vars and rules Currently FFLL does still require all variables to define a rule. As far as...

    2006-01-24 23:06:14 UTC in FFLL - Free Fuzzy Logic Library

  • Followup: RE: problem at the load_from_fcl_file

    Sorry for the delayed reply. The "could be more robust" comment meant that the the EXACT error should be found and returned. If the error is occuring after the line: >> if (!(file_contents.is_open())) Make sure the "file_name" is accessable. Try using the absolute path to the file. If the error is past that in load_from_fcl_file there is something wrong...

    2005-09-04 15:46:18 UTC in FFLL - Free Fuzzy Logic Library

  • Followup: RE: problem at the load_from_fcl_file

    Sorry for the delayed reply. The "could be more robust" comment meant that the the EXACT.

    2005-09-04 15:42:20 UTC in FFLL - Free Fuzzy Logic Library

  • Followup: RE: Combs Method?

    Adding the Combs Method is somewhere on my infinite to-do list... however, since FFLL is open-source anyone can feel free to tackle the challenge! Michael Z.

    2005-08-09 22:10:46 UTC in FFLL - Free Fuzzy Logic Library

  • Followup: RE: DOM for output variables

    dilaylaa, The function you added will not work for OUTPUT variables. They are treated a bit different (as documented) to make FFLL faster. To eliminate extra calculations, FFLL does not calculate the defuzzified output value until it is requested. To get the DOM for an OUTPUT set you'll need to modify your ffll_get_dom() function. Look at the ffll_get_output_value() function and make use...

    2005-08-04 13:07:36 UTC in FFLL - Free Fuzzy Logic Library

  • Followup: RE: DOM for output variables

    There are a few problems with your code: (1) you never load the FCL (2) you don't get a child for the model... you're passing in an initialized variable (childBew) (3) Singleton output should always have a 100% DOM... normally when you use a singleton you want a discrete output value, something like "I'm 33 year old" not "I'm fairly young".

    2005-08-02 15:40:46 UTC in FFLL - Free Fuzzy Logic Library

  • Followup: RE: DOM for output variables

    I quickly put in the function you wrote and get_dom() works for me on the output variable. My guess is the problem is in the FCL... please post that too (or email it to me michaelz at users.sourceforge.net) Also... pay attention to the data types: get_dom() should return DOMType. Your function mixes double and RealType Hope this helps! Michael Z.

    2005-08-02 00:30:44 UTC in FFLL - Free Fuzzy Logic Library

  • Followup: RE: DOM for output variables

    There currently is no function in the API to get the output DOM, *HOWEVER* FuzzyOutVariable inherits from FuzzyVariableBase so you can get to the get_dom() (which gets the DOM per SET). You will just need to add a function to the API. Michael Z.

    2005-07-29 14:25:57 UTC in FFLL - Free Fuzzy Logic Library

  • Followup: RE: aiwisdom in visual c++

    I can run the exe from the CD on WinXP... it's very difficult to say what the problem is without being on your actual machine. My advice is to debug with a bunch of printf()s to narrow down where the problem is. Sorry I can't be more specific! Michael Z.

    2005-07-07 15:47:49 UTC in FFLL - Free Fuzzy Logic Library

About Me

  • 2001-01-23 (9 years ago)
  • 140371
  • michaelz (My Site)
  • Michael Zarozinski

Send me a message