Machine Learning 

Note: Contents are dynamic, and will be updated accordingly as per class discussion. So keep visiting this page.

ML Prerequisites and prework  (Optional: We will not discuss these in detail in the class. As these related with previous semesters' course work. I will also discuss but briefly during the lectures as well. A student who lacks in basics can start self study from here to coverup the deficies of previous semesters related to basic mathematics, statistics, programming concepts. In ML we dont reqiure intensive programming as we usually do in non ML projects, so dont worry much about it. I will teach how to proceed insha'Alah.)

Week 1-4:

• Introduction to Machine Learning

• Linear Regression 

• Classification

• Datacamp course (DC1): Supervised Learning with scikit-learn  (Assignment 1: Course certificate)

  • Lecture Slides

Week 5-8:

• Neural Networks

  • Datacamp course (DC2): Introduction to Deep Learning with PyTorch (Assignment 2: Course certificate)

    • Lecture Slides

  • Datacamp course (DC3): Intermediate Deep Learning with PyTorch (Assignment 3: Course certificate)

    • Lecture Slides

  • Discuss semester Project(s) P1, and P2 (see below for details)

Data (these topics will be discussed at appropriate places throughout the course)

• Working with numerical data 

• Working with categorical data

• Datasets, generalization, and overfitting

Week 9-10: Unpuervised Learning  (Clustering: K-means,Performance evaluation of k-means, Dimensuion Reduction: PCA, ...)

• Datacamp course (DC4): Unsupervised Learning in Python  (Assignment 4: Course certificate)

  • Lecture Slides

• Slides: Dimension Reduction PCA

  • PCA Part 1

  • PCA Part 2

Week 11:  

• Production ML systems 

Week 12:  

• ML Fairness

Week 13:  

• ML Problem Framing

Week 14:  

• Managing ML projects

Week 15-16:  (Projects) 

Visit these projects after week 6, We can have P1 in the class during week 5-8 depending on performance of students.

• P1: Image Classification (Theory & code along project and it can be taught or assigned with the topic of ANN)

• P2: Project: Developing Multi-Input Models For OCR (Prerequisite: Intermediate Deep Learning with PyTorch)

Note: Decision Forests  (If time permits we will discuss otherwise do it yourself after the final exam)

Books and Related Material:

1. ST: Associate AI Engineer for Data Scientists (Datacamp skill track)

   1. 1. 1. Datacamp course (DC1): Supervised Learning with scikit-learn 

         2. Datacamp course (DC2): Introduction to Deep Learning with PyTorch

         3. Datacamp course (DC3): Intermediate Deep Learning with PyTorch

         4. Datacamp course (DC4): Unsupervised Learning in Python 

2. B1: Data Mining a Tutorial Based Primer-richard-j-roiger-2nd-ed (Book)

3. B2: Introduction to Machine Learning with Python- A GUIDE FOR DATA SCIENTISTS (Book)

4. B3: Mastering Machine Learning with scikit-learn, 2nd Edition (Book)

5. B4: Machine-Learning-Yearning-Andrew-Ng-10Sep25  (An introductory book about developing ML algorithms )

6. G1: Machine Learning Crash Course (Google ML crash course)

7. G2: Advance ML Topics (Google ML crash course - advance topics)

8. Deep Learning with PyTorch Cheat Sheet (Self study)

9. PyTorch Tutorial: Building a Simple Neural Network From Scratch (Self study)