Timing

Marimo

Note

code examples are run using marimo notebooks: https://marimo.io/

Simple timing

• using the time.perf_counter() function to set the start point,
• Calculate it again at the end of what you want, and subtract the difference

File I/O

• File I/O is usually the slowest part of your application
• You can see that parquet files work better in general over CSVs

More detailed timing

• Python has the cProfile module that lets you time all the functions and components
• The snakeviz library helps make the cProfile table a bit more readable

Timing in Shiny for R

• Shiny for R has many tools to help you time your `

https://profvis.r-lib.org/

• Profviz example: https://profvis.r-lib.org/examples.html#example-3---profiling-a-shiny-application

Timing in Shiny for Python

• There are no Shiny specific tools for profiling a Shiny for Python app at the moment