Description
In this homework, you will implement a discrimination by regression algorithm for multiclass
classification in Matlab, Python, or R. Here are the steps you need to follow:
1. Read Section 10.7.3 (4th edition) or Section 10.8 (3rd edition) from the textbook.
2. You are given a multivariate classification data set, which contains 195 handwritten
letters of size 20 pixels × 16 pixels (i.e., 320 pixels). These images are from five distinct
classes, namely, A, B, C, D, and E, where we have 39 data points from each class. The
figure below shows five sample figures from each class. You are given two data files:
a. hw02_data_set_images.csv: letter images,
b. hw02_data_set_labels.csv: corresponding class labels.
3. Divide the data set into two parts by assigning the first 25 images from each class to the
training set and the remaining 14 images to the test set.
4. Learn a discrimination by regression algorithm using the sigmoid function for this
multiclass classification problem. You can use the following learning parameters.
eta <- 0.01
epsilon <- 1e-3
5. Draw the objective function values throughout the iterations. Your figure should be
similar to the following figure.
6. Calculate the confusion matrix for the data points in your training set using the
discrimination rule you will develop using the estimated parameters. Your confusion
matrix should be similar to the following matrix.
y_train
y_predicted 1 2 3 4 5
1 25 0 0 0 0
2 0 25 0 0 0
3 0 0 25 0 0
4 0 0 0 25 0
5 0 0 0 0 25
7. Calculate the confusion matrix for the data points in your test set using the parametric
discrimination rule you will develop using the estimated parameters. Your confusion
matrix should be similar to the following matrix.
y_test
y_predicted 1 2 3 4 5
1 13 1 0 0 0
2 1 11 0 0 2
3 0 0 14 0 0
4 0 1 0 14 0
5 0 1 0 0 12
What to submit: You need to submit your source code in a single file (.m file if you are using
Matlab, .py file if you are using Python, or .R file if you are using R) and a short report
explaining your approach (.doc, .docx, or .pdf file). You will put these two files in a single zip
file named as STUDENTID.zip, where STUDENTID should be replaced with your 7-digit
student number.
0 200 400 600 800 1000
0
20
40
60
80
Iteration
Error
How to submit: Submit the zip file you created to Blackboard. Please follow the exact style
mentioned and do not send a zip file named as STUDENTID.zip. Submissions that do not follow
these guidelines will not be graded.
Late submission policy: Late submissions will not be graded.
Cheating policy: Very similar submissions will not be graded.
