ferrifuzz.plot
Quick matplotlib figures for coursework and reports. matplotlib and numpy are
optional: install them with pip install "ferrifuzz[plot]" (or pip install numpy matplotlib).
The module is only imported when you first use it, and without matplotlib a call raises
an ImportError that says what to install.
import matplotlib.pyplot as plt
import ferrifuzz as ff
import ferrifuzz.plot as fp # or use ff.plot.plot_variable(...) directly
Every function draws on ax (a new figure when ax is omitted) and returns the axes,
so you can add titles and labels and save the figure as usual. Functions that need a
value for inputs you are not plotting take fixed= (a dict); inputs not listed there sit
at the middle of their range.
| Function | Draws |
|---|---|
plot_variable(var, ax=None, n=400, legend=True) | every term of a variable |
plot_output(system, inputs, output, ax=None) | how one output is formed for one input row |
plot_response(system, input, output, fixed=None, n=200, ax=None) | output vs one input |
plot_surface(system, x, y, output, fixed=None, n=40, ax=None, cmap="viridis") | 3‑D control surface over two inputs |
plot_variable
All terms of a Variable over [min_value, max_value] (sampled at n points), with a
legend. Interval type‑2 terms are drawn as a shaded band between the lower and upper MFs
(the footprint of uncertainty).
Setup used by the examples on this page: the two‑input tipper
import ferrifuzz as ff
service = ff.Variable("service", 0, 10)
service["poor"] = ff.mf.trapezoid(0, 0, 2, 5)
service["good"] = ff.mf.triangle(2, 5, 8)
service["excellent"] = ff.mf.trapezoid(5, 8, 10, 10)
food = ff.Variable("food", 0, 10)
food["rancid"] = ff.mf.trapezoid(0, 0, 1, 3)
food["delicious"] = ff.mf.trapezoid(7, 9, 10, 10)
tip = ff.Variable("tip", 0, 30)
tip["low"] = ff.mf.triangle(0, 5, 10)
tip["medium"] = ff.mf.triangle(10, 15, 20)
tip["high"] = ff.mf.triangle(20, 25, 30)
system = ff.FuzzySystem()
system.add_input(service)
system.add_input(food)
system.add_output(tip)
system.add_rules([
"IF service IS poor OR food IS rancid THEN tip IS low",
"IF service IS good THEN tip IS medium",
"IF service IS excellent OR food IS delicious THEN tip IS high",
])
ax = fp.plot_variable(service)
ax.set_title("service")
plt.savefig("service.png", dpi=150)
plot_output
The output variable's terms (faint), the crisp result as a dashed vertical line, and:
- Mamdani: the aggregated output set (filled), i.e. what
system.output_setreturns; - interval type‑2: the aggregated footprint of uncertainty and the type‑reduced interval ;
- Sugeno: just the terms and the crisp value.
The quickest way to see why a defuzzifier returned what it did.
fp.plot_output(system, {"service": 7.5, "food": 9.0}, "tip")
plot_response
A 1‑D response curve: output as input sweeps its universe in n steps (computed with
compute_batch).
fp.plot_response(system, "service", "tip", fixed={"food": 5.0})
plot_surface
A 3‑D control surface of output over an n × n grid of inputs x and y, computed
with compute_batch. Pass a 3‑D axes as ax (fig.add_subplot(projection="3d")) to
place it in your own figure.
fp.plot_surface(system, "service", "food", "tip")
plt.show()
Unknown input or output names raise ff.InputError.
Without the helpers
Everything here can be done with plain matplotlib and the core API, which is useful if you want full control over styling:
import numpy as np
xs = np.linspace(service.min_value, service.max_value, 300)
for term in service:
plt.plot(xs, service[term].membership(xs), label=term)
plt.legend()
plt.show()