Dive into advanced Physics-Informed Neural Networks (PINNs) using NVIDIA Modulus with the "NVIDIA Modulus: Advanced Topics" course. Explore topics including I-PINNs, DeepONet, FNO, PINNs for 3D Linear Elasticity Problem, Multi-Domain Calculation, and Geometric Optimization. Learn to write and build ML algorithms to solve PINNs using Nvidia Modulus, including post-processing and pre-processing data with open-source libraries.
All you need is a background in High School Math and basic Python knowledge. From the mathematical foundations to implementation, the course covers solving partial differential equations (PDEs) with PINNs, I-PINNs, DeepONet, FNO, Multi-Domain Calculation, and Geometric Optimization. By the end, you'll be equipped with the skills necessary to tackle these advanced topics confidently.
Whether you're an engineer or a programmer, this course will expand your knowledge of PINNs and advanced topics in NVIDIA Modulus. By the end of NVIDIA Modulus: Advanced Topics, you'll be able to implement PINNs for various problems, including 2D heat sink flow, integration, Darcy problem, 3D Linear Elasticity Problem, Multi-Domain Calculation, and Geometric Optimization for the Heat Exchanger Flow Problem. Join us on this learning journey, and let's explore the world of advanced topics in NVIDIA Modulus together.
NVIDIA Modulus: Advanced Topics Table of Contents:
- Introduction: 2 lectures (11 min)
- Inverse PINNs: 11 lectures (1 hr 36 min)
- Deep Neural Operator (DeepONet): 11 lectures (1 hr 17 min)
- Deep Neural Operator (FNO - Fourier Neural Operator): 10 lectures (59 min)
- 3D Bracket Stress Analysis: 10 lectures (1 hr 44 min)
- PINNs Multi-Domain Calculation: 16 lectures (3 hr 59 min)
- Geometric Optimization using PINNs: 6 lectures (19 min)
Who is this course for?
- Engineers and programmers interested in learning about PINNs
- Those aiming to explore advanced topics in NVIDIA Modulus
