I have a few worker processes for fitting models (running iterative improvement algorithms) and a main process listening to events, If certain events arrive, I will need to terminate the worker processes and retrieve their current model params from the worker process (before the fitting stops/converges)

My thought is to create a shared variable for event, like, setting its value to 1 if certain event happens, and 0 otherwise. So, when worker process checks its value equal to 1, it calls "terminate" in the worker process and return the models params.

This is all I could think of, does anyone have a better idea of how to approach this? Thank you very much!

CodePudding user response：