Explore grain growth modeling and array programming techniques in this 18-minute conference talk from Dyalog '23. Discover how materials science students can benefit from using array programming and APL to solve computational problems. Follow the journey from a simple Python implementation to more advanced solutions using NumPy and APL. Learn about the challenges of modeling grain growth, including addressing unrealistic patterns and singular grain cases. Compare the performance of different programming approaches through benchmarks, and gain insights into embedding APL with Py'n'APL. Understand the practical applications of array programming in materials science and the potential advantages of using APL for computational tasks.
Grain Growth and Array Programming - Modeling Techniques for Materials Science
Dyalog User Meetings via YouTube
Overview
Syllabus
A problem for materials science students
Modelling grain growth
First solution in Python
Partial results reveal the problem with singular grain case
Consider second order neighbours to address unrealistic checkboard pattern
Array solution with NumPy
APL grain growth model
Embed APL with Py'n''APL
Benchmarks comparing performance of APL, Python, and NumPy
Conclusions
Taught by
Dyalog User Meetings
