Lab Session -11
Artificial Intelligence (Comp – 00634)
Lab Session-11
Objective: Understanding the Basics of Image Classification using Deep Neural Networks
(DNN)
A- Outcomes:
After completion of the lab session students will be able:
a. To understand the basics of gray-scale images in MNIST datasets
b. To understand image processing
c. To understand image classification through deep neural network (DNN)
d. To understand the impact of various hyper-parameters of DNN on image
classification accuracy
1
Lab Session -11
B- Lab Tasks:
1- Create a set of classifications for each of the test images and then print the first entry
in the classifications. The output will be list of numbers. Why do you think this is, and
what do these numbers represent?
Write/copy your code here:
2- Let’s now look at the layers in your model. Experiment with different values for the
dense layer with 512 neurons. What different results do you get for loss, training time
etc.? Why do you think that’s the case?
Write/copy your code here:
3- What would happen if you remove the Flatten() layer. Why do you think that’s the
case?
Write/copy your code here:
4- Consider the final (output) layers. Why are there 10 of them? What would happen if
you had a different amount than 10? For example, try training the network with 5.
Write/copy your code here:
5- Consider the effects of additional layers in the network. What will happen if you add
another layer between the one with 512 and the final layer with 10?
Write/copy your code here:
2
Lab Session -11
6- Consider the impact of training for more or less epochs. Why do you think that would
be the case?
Write/copy your code here:
7- Before you trained, you normalized the data, going from values that were 0-255 to
values that were 0-1. What would be the impact of removing that? Here’s the
complete code to give it a try. Why do you think you get different results?
Write/copy your code here:
8- Earlier when you trained for extra epochs you had an issue where your loss might
change. It might have taken a bit of time for you to wait for the training to do that, and
you might have thought ‘wouldn’t it be nice if I could stop the training when I reach a
desired value?’ – i.e. 60% accuracy might be enough for you, and if you reach that
after 3 epochs, why sit around waiting for it to finish a lot more epochs….So how
would you fix that? Like any other program…you have callbacks! Let’s see them in
action.
Write/copy your code here:
9- Write a conclusion of this lab in your own words.
3
Lab Session -11
Write your answer here by hand:
4
Lab Session-11
