Grid Parameter Search¶
Exhaustive search over hyperparameters.
Documentation¶
Evaluates each parameter combination in the set of parameter hypercubes given by each table in the list of parameter tables. Each parameter table specify a hypercube of dimension N where N is the number of columns in the table. Every combination of parameter values in each such cube will be sampled.
Definition¶
Input ports¶
- in-model
Type: modelDescription: in-model- parameter space
Type: [table]Description: param-space- X
Type: tableDescription: X- Y
Type: tableDescription: Y- cross-validation
Type: [(table,table)]Description: cross-validationOptional number of ports: 0–1 (default: 0)
Output ports¶
- results
Type: tableDescription: results- parameters
Type: tableDescription: parameters- out-model
Type: modelDescription: out-model
Configuration¶
- Cross validation splits (cv)
Number of fold in the default K-Fold cross validation. Ignored when cross-validation port is given
Examples¶
The node can be found in:
Implementation¶
- class node_paramsearch.ParameterSearch_Grid[source]