Documentation to learn scikit-fuzzy

[Asejage Poypularis]

Hi,

for my master thesis i have to learn Pyhton and scikit-fuzzy as quickly as possible  On You-tube i found a video from Mr. Warner. He shows in the minute 5:29 a web-page, that seems like a tutorial. But I'm missing the tutorial about scikit-fuzzy. How and where did you learned it? 

Best regards,

Asejage

[Asejage Poypularis]

Nobody can help me? :( I just need a source, where i can learn scikit-fuzzy. 

Mr. Warner, can you tell me the source from the video : https://www.youtube.com/watch?v=qUQf1JxnTnY , please? or any source, where i can learn scikit-fuzzy?

Thank you!

[JDWarner]

Hi Asejage,

I've written about 3 possible responses and scrapped all of them. This is not a simple question. Long story short: the demo in my SciPy 2013 presentation is a limited demo contained in an IPython Notebook. It is not even close to a full tutorial for the package; it touches on the merest fraction of the capabilities of scikit-fuzzy. However, if you want to peruse it, the notebook is hosted in an un-merged branch on my personal fork of scikit-fuzzy: https://github.com/JDWarner/scikit-fuzzy/blob/scipy2013/scikit-fuzzy_demo.ipynb

We are working on building additional documentation and a website, but presently the package is documented mainly in individual function docstrings. Within IPython, you can explore the available functions with tab completion and inspect these docstrings with an additional question mark, similar to other Python packages. Which brings me to the next point: you will need to know, at least the basics of the scientific Python ecosystem. Most importantly, nearly every function in scikit-fuzzy uses the NumPy ndarray structure for input and output. This is functionally similar to a MATLAB array, and if you are familiar with MATLAB this reference may be of assistance to get you started: http://wiki.scipy.org/NumPy_for_Matlab_Users

Generally speaking, I would not recommend learning scikit-fuzzy; rather, I would recommend learning the principles of fuzzy logic and Python for scientific use in parallel and you will be able to, organically, find the appropriate commands and workflows for implementing what you want to do in Python using scikit-fuzzy. The design of the package exposes the vast majority of the features using a functional interface in the base namespace, so once installed a simple `import skfuzzy as fuzz; dir(fuzz)` will list the available functions. They are descriptively named, so as you learn fuzzy logic the package will grow with you. The design right now builds out most of the necessary underlying framework necessary for fuzzy logic, enabling higher level applications.

I hope that points you in the right direction,

Josh Warner

[Stavros Anastasiadis]

 Iam also starting using scikit- fuzzy so the following examples might help you (they are not full ready yet , but its a try  )

1. Tipping Problem 

2. Fuzzy Math

[Lara]

Thank you very much for your both answers!! They will help me in any case.

Best regards! 

[JDWarner]

For posterity, if anyone comes here from Google:

We now have package documentation available online, and the example gallery is an excellent place to begin!

Regards,

Josh