COMP 536: Computational Modeling for Scientists

Spring 2026

Welcome to COMP 536

COMP 536 teaches you to think computationally — to translate scientific questions into code, build numerical models, and extract insights from data. Whether you’re simulating physical systems, analyzing experimental results, or processing large datasets, this course builds the practical skills you’ll use throughout your research career.

You’ll learn Python from the ground up, then build to powerful libraries like NumPy, Matplotlib, SciPy, and Pandas. Along the way, you’ll develop professional practices: version control with Git, reproducible workflows, and code that others (including your future self) can understand.

By the end, you’ll be able to solve differential equations numerically, fit models to data, create publication-quality figures, and write modular, maintainable scientific code.

Instructor

Dr. Anna Rosen

Physics 239

Fridays 11:00 am–12:00 pm (and by appointment)

alrosen@sdsu.edu

---

Note Start Here

1. Read the Syllabus to understand course policies and expectations.
2. Complete the Getting Started tutorials to set up your environment.
3. Submit all assignments on Canvas.

Course Materials

Course Info

Syllabus, policies, and course logistics.

• Syllabus
• AI Policy

Getting Started

Set up your computational environment and learn essential tools.

• Command Line Introduction
• Software Setup
• Git Introduction

Scientific Computing with Python

Build your Python foundation for computational science.

Python Fundamentals

• Ch 1: Computational Environments
• Ch 2: Python as Your Calculator
• Ch 3: Control Flow & Logic
• Ch 4: Data Structures
• Ch 5: Functions & Modules
• Ch 6: OOP Fundamentals

The Learnable Universe

Use modern scientific computing tools to rebuild simulations, generate trustworthy training data, and train fast surrogate models for the final project.

Computing the Universe with JAX

• Overview: Computing the Universe with JAX
• Part 1: Conceptual Foundations
• Part 2: Core Transformations
• Part 3: Physics Applications

Machine Learning for Emulators

• Overview: Introduction to Machine Learning
• Part 1: Neural Networks and Scientific Emulation
• Part 2: Uncertainty Quantification

How to Succeed

1. Write code every day. Programming is a skill — practice builds fluency.
2. Read error messages carefully. They’re trying to help you.
3. Use version control. Commit early, commit often.
4. Ask early. Office hours exist for a reason.

Tip Getting Help

• Office hours: Fridays 11:00 am–12:00 pm (and by appointment) in Physics 239
• Canvas discussion: Post questions others might share