I always care about performance when I write code. I also care about redability. How to deal with this tradeoff? Sometimes I prefer redability if the decrease in performance is not high enough, but it's not comfortable to look for the performance of each operation. So I decided to write a suite to evaluate the differences between different options, along with a set of handy notes sorted by topic.
timeperf is written to not need more libraries than necessary. If you want to
perform the entire set of tests, you can install the dependencies listed in
requirements.txt file by running:
$ pip install -r requirements.txt
Or you could create a virtual environment and install them on it:
$ mkvirtualenv timeperf
(timeperf) $ pip install -r requirements.txt
To run the code you need to create one function for each operation that you
want to perform. Then you pass these functions as parameters for the Runner
object to obtain the metrics.
TODO
This is an open source project, so you are invited to add the measures you consider useful. The idea is to have a unified place to quickly search for the best ways to perform different operations.
- Divide each measurement based on Python version?