search

mle-infrastructure / mle-toolbox

Lightweight Tool to Manage Distributed ML Experiments 🛠
https://mle-infrastructure.github.io/mle_toolbox/toolbox/
MIT License
3 stars 1 forks source link

Command line argument for resource to run on #15

Closed RobertTLange closed 3 years ago

RobertTLange commented 3 years ago

Add a command line argument --resource {local, sge, slurm, gcp} that lets you decide where to run on. If this is not given resort back to asking user through the CLI.

RobertTLange commented 3 years ago

This mostly already existed but only for remote case. I added the local option (see 05f5dd9a3b97407945671b4497578e451b0ea680). So now we can get rid of most of the CLI input with the following arguments:

run-experiment ode/ode_search.yaml --no_protocol --no_welcome --resource_to_run local

or in short:

run-experiment ode/ode_search.yaml -np -nw -resource local