Hi everyone, i am developing a pipeline of trainin...
# questions
Hi everyone, i am developing a pipeline of training which has some parameters set by default. I have a question about what are the best practices for running multiple times the same pipeline but with multiple parameters. I am currently using a bash script that runs “kedro run” with the flag “--params”. I am doing this because i want to register each experiment. However it does not fell quite right and i was wondering if it is possible to run multiple times a pipeline varying the parameters via a hook implementation or should i use a modular pipeline and register all the parameters in the yaml file. (i think this solutions is not quite practical ) but i would like to know if there is a “Better” way of doing these kind of things
👍 1
I think it’s better to have the run separately as you need to have 1 run as 1 experiment. This depends on what kind of experiment tracking tools you are using.
If you are using the Python API
, it should be easy to read parameters and go through a loop to execute all the runs. This is almost equivalent to run a bash script.
Ok thanks for your help 😄