Syllabus
Scannable course roadmap: goals, structure, assessment, and policies.
Course reference
Overview
Foundations of Data Science is an undergraduate course offered by the Department of Electrical Engineering at Sharif University of Technology. The course combines theoretical foundations with hands-on learning, emphasizing practical skills, computational thinking, and the development of end-to-end machine learning workflows.
The course covers essential topics in data analysis and statistical learning before progressing to contemporary approaches in deep learning, generative modeling, and modern AI systems. Through a combination of lectures, practical exercises, and project-based work, students develop the knowledge and skills needed to apply machine learning methods to real-world problems.
Instructors are Dr. Babak Hossein Khalaj and Dr. Amir Hossein Saberi.
Learning objectives
By the end of the course, you should be able to:
- Use core data-science tools and data representations confidently.
- Apply probability, statistics, and classical ML methods with sound evaluation.
- Explain modern neural architectures, transformers, and generative models at a foundational level.
- Build end-to-end ML workflows from data processing through monitoring.
Prerequisites
Recommended background before the first weeks:
- Programming familiarity (Python preferred).
- Undergraduate mathematics: linear algebra, calculus, and basic probability.
- Confirm any formal prerequisites with course staff if unsure.
Course structure
The semester follows four arcs: foundations and data tooling; statistical learning; deep learning and generative models; then ML dataflow modules and the multi-phase project. See Lectures for the ordered session list.
Foundations & data tooling
Introduction, data models, databases, visualization, and ML dataflow — processing.
Statistical learning
Probability and statistics, regression, supervised and unsupervised learning, causality, and model evaluation.
Deep learning & generative models
Neural networks, transformers and small language models, generative and diffusion models, and modern time-series modeling.
ML dataflow & project
Implementation, pipelines, monitoring, and the multi-phase course project through final presentation.
Full session order: Lectures
Evaluation / grading
Your grade is built from a 20-point core (homework, course project, final exam, and TA quizzes), plus bonus opportunities on the project and in class. The table below matches the official grading policy.
| Component | Points | Details |
|---|---|---|
| Homework | 7.5 | Six assignments (Homework 1–6), including the time-series problem set. Deadlines and responsible TAs are on the Assignments page. |
| Course project | 4.5 (+1 bonus) | Multi-phase course project (Phases 1–3) and final presentation. Up to one additional bonus point may apply as announced in class. |
| Final exam | 5 | End-of-term exam covering the full course; emphasis by module as announced by instructors. |
| Teaching assistant quizzes | 3 | Four quizzes led by teaching assistants; topics and dates follow lecture and workshop coverage. |
| Class quiz | Bonus | In-class quizzes with floating bonus points. Attend and participate to earn credit toward your grade. |
Deadlines and owners: Assignments
Policies
Follow in-class announcements for attendance, late work, academic integrity, and preferred contact channels. Prefer the responsible TA listed on each assignment when asking about a specific homework or project phase; use the Staff page for emails and Telegram when published. Policy details will be expanded here once instructors finalize them.
Where to go next
Use the syllabus as the roadmap; open these pages for day-to-day work.
- LecturesSession timeline, presenters, lecture materials
- WorkshopsTA-led sessions, videos, notebooks
- AssignmentsHomeworks, project phases, responsible TAs
- ScheduleCalendar dates when published
- StaffInstructors, contact channels