Create a computer vision system using decision tree algorithms to solve a real-world problem : [Activity] Classifying Im

The video tutorial covers the process of working with a dataset containing 10 categories of images, including airplanes, automobiles, and more. It explains how to display images using matplotlib, normalize data by scaling pixel values, and create a Convolutional Neural Network (CNN) model. The tutorial also discusses training the model over 10 epochs, evaluating its performance, and identifying areas for improvement. The model achieves a 62% accuracy rate, with suggestions for further tuning and optimization.

What is one of the challenges mentioned in classifying the images in the dataset?

• The images are too large.

• The images are too colorful.

• The images are in black and white.

• The images are grainy and small.

Why is it necessary to normalize the image data before feeding it into the CNN?

• To increase the image size.

• To convert the images to grayscale.

• To scale the pixel values to a range of 0 to 1.

• To change the image format.

What is the purpose of using a dropout layer in the CNN model?

• To prevent overfitting by randomly dropping units.

• To increase the number of neurons.

• To enhance the color of images.

• To convert images to a different format.

How many epochs were used to train the CNN model in the video?

• 20 epochs

• 15 epochs

• 10 epochs

• 5 epochs

What was the test accuracy achieved by the CNN model after training?

• 72%

• 62%

• 82%

• 52%

What is one reason the model misclassified some images?

• The images were too colorful.

• The images were too large.

• The model used the wrong algorithm.

• The model was not trained long enough.

What is a potential next step to improve the model's performance?

• Tune the model's topology and run for more epochs.

• Use a different dataset.

• Convert images to black and white.

• Increase the image size.