Lead: Santiago Nunez-Corrales
Members: Minu Mathew Todd Nicholson Vismayak Mohanarajan
In this Focus Group, we will focus on gathering practice and experience around the following questions:
- Which factors determine when to use Pandas and when to use Numpy?
- How do both packages represent data and with which trade-offs?
- What are adequate ways to access SciPy functions to preserve performance?
- What can be done when alternatives provided by these packages are exhausted and performance still needs to improve?
- How large are the gains of using Cython, Numba and FFIs vs plain Python code?
- Concerns at large: How to gauge, project-wise, the balance between maintainability and performance?
We will pursue the following outcomes:
- A series of recommendations and good practices that serve the needs of projects, teams and professionals across NCSA
- Short (30-min) hands-on sessions that demonstrate specific pain points and possible solutions
- Actions that foster communication between NCSA software developers and open-source package maintainers