Scribd
Upload
23 views5 pages

AI Practical Modul1 v1

The document outlines a practical module for students to build a simple image classification model using Machine Learning with the CIFAR-10 dataset. It includes steps for setting up the environment, training the model, evaluating its performance, and making predictions, along with discussions on key concepts. Expected learning outcomes include understanding AI model training, evaluation, and improvement strategies.

Uploaded by

slebenasep
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
23 views5 pages

AI Practical Modul1 v1

The document outlines a practical module for students to build a simple image classification model using Machine Learning with the CIFAR-10 dataset. It includes steps for setting up the environment, training the model, evaluating its performance, and making predictions, along with discussions on key concepts. Expected learning outcomes include understanding AI model training, evaluation, and improvement strategies.

Uploaded by

slebenasep
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

Your_name_your_StudentID .

Image Classification
Using Machine Learning
Date: 24/02/2025

Practical Module: Image Classification Using Machine Learning

Course: Advanced AI Applications in Smart Systems

Week: Basic AI Practical Session

IF420 . Miratul Khusna Mufida


Your_name_your_StudentID . 2

Objective
Students will build a simple image classification model using Machine Learning. They will
learn to train an AI model to classify images from the CIFAR-10 dataset and evaluate its
performance.

1.​Materials Required
✅ Google Colab (or Jupyter Notebook)​
✅ Python Libraries: TensorFlow, Keras, NumPy, Matplotlib​
✅ CIFAR-10 Dataset (Preloaded in Keras)

2.​Setup Google Colab & Import Libraries


Open Google Colab and create a new notebook. Then, run the following code:

3.​Load the Dataset


We use the CIFAR-10 dataset, which contains 60,000 images in 10 classes.
Your_name_your_StudentID . 3

4.​Visualize Sample Images

Discussion: What patterns do students observe? How does AI recognize these images?

5.​Build a Simple Machine Learning Model

Discussion:

●​ What do Conv2D and MaxPooling2D layers do?


●​ Why do we use softmax in the output layer?
Your_name_your_StudentID . 4

6.​Train the Model

This step may take a few minutes. Monitor loss and accuracy trends.

7.​Evaluate Model Performance

Discussion:

●​ What does accuracy mean in classification?


●​ How can we improve this model?

8.​Make Predictions on New Images


Your_name_your_StudentID . 5

Discussion:

●​ Are the predictions correct?


●​ What happens when the AI makes a wrong prediction?

9.​Wrap-Up Discussion
1.​ Key Learnings:​

○​ Basics of training an AI model for image classification


○​ How to evaluate an AI model
○​ The role of convolutional layers in image recognition
2.​ Challenges & Improvements:​

○​ What were the difficulties?


○​ How can we improve accuracy? (e.g., data augmentation, tuning
hyperparameters)

10. Assessment (Homework)


1.​ Modify the model to use more epochs (e.g., 10 epochs). Observe the accuracy.
2.​ Try using a different dataset (e.g., MNIST for digit classification).
3.​ Research and explain what data augmentation is and how it improves AI models.

11. Expected Learning Outcomes


✅ Understand how an AI model learns from data​
✅ Train and evaluate an image classification model​
✅ Identify ways to improve AI performance

12. Next Steps (For Advanced Students)


●​ Experiment with different architectures (e.g., ResNet, MobileNet)
●​ Implement data augmentation for better generalization
●​ Apply AI to other datasets (e.g., handwriting recognition, facial recognition)

You might also like