Open betatim opened 8 years ago
Some previous discussion on this is here: #594 asv
is more than I had in mind 😀 Initial thought was just to collect a few snippets that are being used for benchmarking and organise them a bit like the unit tests so that you can run them without having to think too much (vs freestyling for each benchmark).
Following on from #1436
Create
benchmarks/
as a collection of small short performance benchmarks (using the py.test technology?) for CPU and/or memory performance. Having a shared collection of these benchmarks makes it easier for everyone to use them (and agree on what was wrong) when they are modifying code that they think is performance critical.1436 links to a gist that benchmarks hashing of kmers and converting the hashes back to python integers. Small variation of this script can be used to benchmark just the hash function.