Leverage ray for hyper parameter tuning

nsenno-dbr / many-model-forecasting

Bootstrap your large scale forecasting solution on Databricks with Many Models Forecasting (MMF) Project.

Other

0 stars 0 forks source link

Leverage ray for hyper parameter tuning #6

Open nsenno-dbr opened 1 month ago

nsenno-dbr commented 1 month ago

Currently there is some Hyper parameter optimization using optuna. I believe we can increase the currency of this search using Ray on Spark

ryuta-yoshimatsu commented 1 month ago

This has been on the todo for a while. It's also related to scaling the global model training beyond multinode (currently only a single node is supported). I took a stab at it a few times but failed, but would love to have this working.