micropython-scikit-fuzzy is a fuzzy logic toolkit ported from Python's SciPy scikit-fuzzy toolkit.
The goals of micropython-scikit-fuzzy are:
- To provide the community with a robust toolkit of independently developed and implemented fuzzy logic algorithms
- To increase the attractiveness of scientific MicroPython/CPython as a valid alternative to closed-source options.
- To allow experimenting with embedded control systems without installing software or using cloud based services.
Important notice from scikit-fuzzy: Please cite 
https://github.com/JorgeGMarques/micropython-scikit-fuzzy
The documentation of the scikit-fuzzy library can be found here: https://scikit-fuzzy.github.io/scikit-fuzzy
micropython-scikit-fuzzy shares exactly the same methods, but some examples are tuned down to run in constrained systems.
There are many tools in scikit-fuzzy, the list below shows which modules have been ported:
- cluster/_cmeans.py
✅ Done - cluster/normalize_columns.py
✅ Done - defuzzify/defuzz.py
✅ Done - membership/generatemf.py
✅ Done - fuzzymath/fuzzy_ops.py
👷 Doing - fuzzymath/fuzzy_logic.py
⏰ ToDo - fuzzymath/_continuous_to_discrete.py
⏰ ToDo - intervals/intervalops.py
⏰ ToDo - filters/fire.py
⏰ ToDo - control/antecedent_consequent.py
⏰ ToDo - control/controlsystem.py
⏰ ToDo - control/antecedent_consequent.py
⏰ ToDo - control/fuzzyvariable.py
⏰ ToDo - control/rule.py
⏰ ToDo - control/state.py
⏰ ToDo - control/term.py
⏰ ToDo - control/visualization.py
⏰ ToDo
Examples:
- plot_cmeans.py
✅ Done - plot_defuzzify.py
✅ Done - plot_control_system_advanced.py
⏰ ToDo - plot_tipping_problem.py
❌ legacy, won't port - plot_tipping_problem_newapi.py
👷 Doing
You can use BIPES to easily program MicroPython boards and BIPES's Databoard to visualize/plot the fuzzy computations.
The examples have been changed to automatically plot in the BIPES platform, replacing matplot.
scikit-fuzzy's public chat room on Gitter.im
scikit-fuzzy's Google Groups mailing list
http://groups.google.com/group/scikit-fuzzy
BIPES discussion forum https://github.com/BIPES/BIPES/discussions
MicroPython-Scikit-Fuzzy depends on
- MicroPython > v1.17
- micropython-ulab > v3.3.6
Therefore, MicroPython-Scikit-Fuzzy requires a embedded device (or emulated) runnig MicroPython with ulab module.
If you need help compiling the custom firmware, check out ulab's readme, (for the ESP-32 I recommend using this script)
Please read LICENSE.txt in this directory.
It should be noted that Matlab rounds incorrectly. The IEEE standard (which is how this package behaves) requires rounding to the nearest EVEN number if exactly between, e.g. 1.5 --> 2; 2.5 --> 2; 3.5 --> 4; 4.5 --> 4, etc. This minimizes systematic rounding error. Thus, if re-implementing algorithms from Matlab code, slight inconsistencies in rounded results are expected. These are not bugs, and will not be fixed.