search

HIPS / Spearmint

Spearmint Bayesian optimization codebase
Other
1.54k stars 329 forks source link

Spearmint with SQlite for Quick Deployment in Jupyter Notebooks #61

Open wielandbrendel opened 8 years ago

wielandbrendel commented 8 years ago

I'd love to use Spearmint (it is an awesome package!) but often the overhead of setting up a MongoDB server porting my Jupyter Notebook into a runnable python script is too much. Is there any way to run Spearmint in a similar fashion to hyperopt? Maybe with SQlite as a backend?

mgelbart commented 8 years ago

At the moment MongoDB is the only supported database for storing results in Spearmint. If we let the user call Spearmint rather than Spearmint calling your function (see #82) it would've been easier for you to plug in whatever database you want. But with the current set-up you would need to dig into the code a bit to support this. Basically, you'd need to add another database to spearmint/utils/database that implements abstractdb and then use that one instead.