Scribd
Upload
2 views2 pages

Lab 1

The document outlines a tutorial on neural networks, focusing on the decision boundary and linear discriminant functions. It includes practical exercises using MATLAB and Python to implement a simple neural network, Gaussian membership functions, and perceptron learning rules. Additionally, it covers data visualization techniques using the Fisher iris dataset.

Uploaded by

Nidhin Scaria
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
2 views2 pages

Lab 1

The document outlines a tutorial on neural networks, focusing on the decision boundary and linear discriminant functions. It includes practical exercises using MATLAB and Python to implement a simple neural network, Gaussian membership functions, and perceptron learning rules. Additionally, it covers data visualization techniques using the Fisher iris dataset.

Uploaded by

Nidhin Scaria
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

tutorial

1. Consider a problem with two features (x1 and x2).

The equation of the decision boundary is

y = x1 + x2 − 1.5 or y = wx + b . This is called ‘linear discriminant function’ in statistical studies.

Or the output y = f ( x, w, b) , where x is the input, w and b are parameters.

w is called weights in neural network terminology and b is called the bias. –b is called threshold.

Note that the weight vector decides the orientation of the decision surface and bias decides the distance from the
origin.

In classification problem output y is one of a number of discrete classes or categories. But in a regression problem,
output y represents the values of continuous variables.

Then we studied, how we can solve the problem in a ‘neural network’ way

The second figure is a neural work with a single neuron. What does the neuron do? It sums up all the inputs then
produces an output.

Now copy the following lines into a matlab file and run (or create code using python). This is your first neural network!
x=input('x=');
sum=x(1)+x(2);
if sum>=1.5
out=1;
else out=0;
end
if out==1
disp('diseased')
else
disp('no disease')
end
Then write in the command window x=[.6 .7]

2. Find the output of the following perceptron neural network for the inputs [1 1] and [1 0]

2
 x −c 
−0.5 
  
3. Gaussian membership function is defined by c and σ. y=e

Create a Gaussian membership function with centre as 40 and spread 10 (x varies from 0 to 100)

a. Plot the above function.

b. Create a general function for the Gaussian function.

4. Study the command newp in matlab and solve examples using this command (or study a possible
toolbox in python, use keras). You can solve OR gate.

5. Visualization of data: Load fisheriris data and plot the data in 2-D and 3-D

load fisheriris
gscatter(meas(:,1), meas(:,2), species,'rgb','osd');
xlabel('Sepal length');
ylabel('Sepal width');

6. Write a MATLAB/Python function for perceptron learning rule (PLR)

7. Write a MATLAB/Python function for gradient descent learning rule

You might also like