ResearchTools 0.0.1

A package for various tools that can be useful in scientific research projects.

ResearchTools

A package for various tools that can be useful in scientific research projects. Two design principles drive this project.

1. Heavy emphasis on using functions/closures, procedures are typically more straightforward to re-purpose than Classes/Objects.

   • Closures typically serve a single purpose, and may expose 1-2 other fields for manipulating internal variables.
     • Keep it Simple and Modular, memory is in no shortage these days but CPU cycles are always costly.
   • Numpy ND-arrays for anything that will be used for number-crunching.
   • When more structure is needed, we use the base Python datatypes as much as possible.
     • Lists for mutable integer-indexed Iterables, Tuples for immutable.
     • Dicts for key-value mappings.

2. Everything should be as fast as possible, targetting high performance computing.

   • Numba-accelerated and vectorized functions where it makes sense.
   • Memoization module for caching the results of expensive functions.