Performance deep dive

In this article, we will find out how to most efficiently complete the Raindrops exercise.

The introduction page lists seven idiomatic approaches to this exercise:

1. Using if statements and + to assemble a return string
2. Exploiting "Truthy" and "Falsy" values in ternary checks and an f-string
3. Using one or more sequences, a loop, and an f-string
4. Using one or more sequences and a generator-expression within join().
5. Using a dict of {factors : sounds} for lookup and join()
6. Using itertools.compress()
7. Using functools.reduce()

For our performance investigation, we'll also include an 8th approach that uses Python 3.10's structural pattern matching.

Benchmarks

To benchmark these functions, we wrote a small benchmarking script using the timeit module along with third-party libraries numpy and pandas.

Keep in mind that all these approaches are very fast, and that benchmarking at this granularity can be unstable. That caveat in mind, the two if-statement based approaches benchmark fastest, and the approach using functools.reduce() was the slowest.

The 'dictionary with join' approach came in 5th, with the 'recommended dictionary' approach in 6th, though it can be argued that the slowdown for either dictionary approach is justified by the increased readability and maintainability.

Measurements were taken on a 3.1 GHz Quad-Core Intel Core i7 Mac running MacOS Ventura. Tests used timeit.Timer.autorange(), repeated 3 times. Time is reported in seconds taken per string after calculating the 'best of' time. The timeit module docs have more details, and note.nkmk.me has a nice summary of methods.