david-thrower / cerebros-core-algorithm-alpha

The Cerebros package is an ultra-precise Neural Architecture Search (NAS) / AutoML that is intended to much more closely mimic biological neurons than conventional neural network architecture strategies.
Other
27 stars 4 forks source link

Add some abstractions for basic problems #73

Open david-thrower opened 1 year ago

david-thrower commented 1 year ago

Add an ancillary module or class that loads data like the notebook and searches a problem - specific presumed feasible space.

Dependencies: Fix issues with race conditions with past attempts at making hyperparameter tuning self-contained: hold to #74

Possible limitations:

May be infeasible for images and text, as these may depend on true distributed training.