neural network tutorial 3
TRANSCRIPT
-
8/17/2019 neural network tutorial 3
1/32
BITS PilaniDubai Campus
BITS F312 NNFL
-
8/17/2019 neural network tutorial 3
2/32
BITS PilaniDubai Campus
Dr.V.KALAICHELVI.,ASST.PROF/EEE
-
8/17/2019 neural network tutorial 3
3/32
BITS Pilani, Dubai Campus
● S-shaped function
● Z-shaped function
● Triangular Membership Function
● Trapezoidal Membership Function
● Gaussian Distribution Function
● Pi function
Various Types of Membership Function
V.Kalaichelvi., Asst.Prof/EEE
-
8/17/2019 neural network tutorial 3
4/32
BITS Pilani, Dubai Campus
S shaped membership Function
V.Kalaichelvi., Asst.Prof/EEE
0 , for x a
2[(x-a) / (c-a)]2 , for a x b
S(x, a, b, c) = 1- 2[(x-c) / (c-a)]2 , for b < x c
1 , for x c
a b c
Figure S-shaped Membership Function
-
8/17/2019 neural network tutorial 3
5/32
BITS Pilani, Dubai Campus
Program -S shaped membership Function
V.Kalaichelvi., Asst.Prof/EEE
x = 0:0.02:1;
y = smf(x,[0.2 0.45]);
plot(x,y)xlabel('smf, P=[0.2 0.45 0.65]')
ylim([-0.05 1.05])
-
8/17/2019 neural network tutorial 3
6/32
BITS Pilani, Dubai Campus
S shaped membership Function
V.Kalaichelvi., Asst.Prof/EEE
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
smf, P=[0.2 0.45 0.65]
M e m b e r s h i p f u n c t i o n
-
8/17/2019 neural network tutorial 3
7/32BITS Pilani, Dubai Campus
V.Kalaichelvi., Asst.Prof/EEE
S shaped membership Function
-
8/17/2019 neural network tutorial 3
8/32BITS Pilani, Dubai Campus
V.Kalaichelvi., Asst.Prof/EEE
Program for S shaped MF
-
8/17/2019 neural network tutorial 3
9/32BITS Pilani, Dubai Campus
V.Kalaichelvi., Asst.Prof/EEE
Program for S shaped MF
-
8/17/2019 neural network tutorial 3
10/32BITS Pilani, Dubai CampusV.Kalaichelvi., Asst.Prof/EEE
-
8/17/2019 neural network tutorial 3
11/32BITS Pilani, Dubai CampusV.Kalaichelvi., Asst.Prof/EEE
x = 0:0.02:1;y = zmf(x,[0.25 0.45]);
plot(x,y)
xlabel('zmf, P=[3 7]')
ylim([-0.05 1.05])
Program
-
8/17/2019 neural network tutorial 3
12/32BITS Pilani, Dubai CampusV.Kalaichelvi., Asst.Prof/EEE
Graphical Representation
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
X: 0.26
Y: 0.995
zmf, P=[3 7]
-
8/17/2019 neural network tutorial 3
13/32BITS Pilani, Dubai CampusV.Kalaichelvi., Asst.Prof/EEE
x = 0:0.1:10;
y = zmf(x,[3 7]);
plot(x,y)xlabel('zmf, P=[3 7]')
ylim([-0.05 1.05])
Program –Z Membership Function
-
8/17/2019 neural network tutorial 3
14/32BITS Pilani, Dubai CampusV.Kalaichelvi., Asst.Prof/EEE
Graphical Representation
-
8/17/2019 neural network tutorial 3
15/32BITS Pilani, Dubai CampusV.Kalaichelvi., Asst.Prof/EEE
-
8/17/2019 neural network tutorial 3
16/32BITS Pilani, Dubai CampusV.Kalaichelvi., Asst.Prof/EEE
x = 0:10:100;
y = trimf(x,[10 50 90]);
plot(x,y)
xlabel('trimf, P=[10 50 90 ]')ylim([-0.05 1.05])
Program
-
8/17/2019 neural network tutorial 3
17/32BITS Pilani, Dubai Campus
V.Kalaichelvi., Asst.Prof/EEE
Graphical Representation
0 10 20 30 40 50 60 70 80 90 100
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
trimf, P=[10 50 90]
M e m b e r s h i p F u n c
t i o n
-
8/17/2019 neural network tutorial 3
18/32BITS Pilani, Dubai CampusV.Kalaichelvi., Asst.Prof/EEE
-
8/17/2019 neural network tutorial 3
19/32BITS Pilani, Dubai CampusV.Kalaichelvi., Asst.Prof/EEE
Program for Trapezoidal
x = 0:0.1:10;
y = trapmf(x,[1 5 7 8]);
plot(x,y)
xlabel('trapmf, P=[1 5 7 8]')
ylabel('Membership Function')
ylim([-0.05 1.05])
title('Program for Trapezoidal MF')
grid0 1 2 3 4 5 6 7 8 9 10
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
trapmf, P=[1 5 7 8]
M e m b e r s
h i p F u n c t i o n
Program for Trapezoidal MF
-
8/17/2019 neural network tutorial 3
20/32BITS Pilani, Dubai Campus
V.Kalaichelvi., Asst.Prof/EEE
-
8/17/2019 neural network tutorial 3
21/32BITS Pilani, Dubai CampusV.Kalaichelvi., Asst.Prof/EEE
x = 0:0.1:20;
y = gaussmf(x,[2 5]);
plot(x,y)
xlabel('gaussmf, P=[2 5]')
Program for gaussian
Membership Function
0 2 4 6 8 10 12 14 16 18 200
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
X: 5
Y: 1
gaussmf, P=[25]
-
8/17/2019 neural network tutorial 3
22/32BITS Pilani, Dubai CampusV.Kalaichelvi., Asst.Prof/EEE
-
8/17/2019 neural network tutorial 3
23/32
BITS Pilani, Dubai Campus
This spline-based curve is so named because of its Πshape.
The membership function is evaluated at the points
determined by the vector x.
The parameters a and d locate the "feet" of the curve,while b and c locate its "shoulders.“
The membership function is aproduct smf and zmf membership functions.
Pi Function
V.Kalaichelvi., Asst.Prof/EEE
http://www.mathworks.com/help/fuzzy/smf.htmlhttp://www.mathworks.com/help/fuzzy/zmf.htmlhttp://www.mathworks.com/help/fuzzy/zmf.htmlhttp://www.mathworks.com/help/fuzzy/smf.html
-
8/17/2019 neural network tutorial 3
24/32
BITS Pilani, Dubai CampusV.Kalaichelvi., Asst.Prof/EEE
-
8/17/2019 neural network tutorial 3
25/32
BITS Pilani, Dubai Campus
-
8/17/2019 neural network tutorial 3
26/32
BITS Pilani, Dubai CampusV.Kalaichelvi., Asst.Prof/EEE
-
8/17/2019 neural network tutorial 3
27/32
BITS Pilani, Dubai CampusV.Kalaichelvi., Asst.Prof/EEE
-
8/17/2019 neural network tutorial 3
28/32
BITS Pilani, Dubai CampusV.Kalaichelvi., Asst.Prof/EEE
-
8/17/2019 neural network tutorial 3
29/32
BITS Pilani, Dubai CampusV.Kalaichelvi., Asst.Prof/EEE
Graphical Representation
-
8/17/2019 neural network tutorial 3
30/32
BITS Pilani, Dubai CampusV.Kalaichelvi., Asst.Prof/EEE
x = (0:0.1:10)';
y1 = trapmf(x, [2 3 7 9]);
y2 = trapmf(x, [3 4 6 8]);
plot(x, [y1 y2]);
ylim([-0.05 1.05])
grid
0 2 4 6 8 10
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Two Trapezoidal MF in onefigure window
-
8/17/2019 neural network tutorial 3
31/32
BITS Pilani, Dubai CampusV.Kalaichelvi., Asst.Prof/EEE
x = (0:0.1:10)';y1 = gaussmf(x, [0.5 5]);
y2 = gaussmf(x, [1 5]);
y3 = gaussmf(x, [2 5]);
y4 = gaussmf(x, [3 5]);
subplot(211); plot(x, [y1 y2 y3 y4]);
y1 = gaussmf(x, [1 2]);
y2 = gaussmf(x, [1 4]);
y3 = gaussmf(x, [1 6]);
y4 = gaussmf(x, [1 8]);
subplot(212); plot(x, [y1 y2 y3 y4]);
Program with Subplotcommand
0 2 4 6 8 10
0
0.2
0.4
0.6
0.8
1
0 2 4 6 8 10
0
0.2
0.4
0.6
0.8
1
-
8/17/2019 neural network tutorial 3
32/32
V Kalaichelvi Asst Prof/EEE
Use of Subplot
SUBPLOT Create axes in tiled positions.
H = SUBPLOT(m,n,p), or SUBPLOT(mnp), breaks the Figure window
into an m-by-n matrix of small axes, selects the p-th axes for
the current plot, and returns the axes handle. The axes are
counted along the top row of the Figure window, then the second
row, etc. For example,
SUBPLOT(2,1,1), PLOT(income)
SUBPLOT(2,1,2), PLOT(outgo)
plots income on the top half of the window and outgo on thebottom half. If the CurrentAxes is nested in a uipanel the
panel is used as the parent for the subplot instead of the
current figure.