Grid Search Combinations Calculator

Work out grid search combinations instantly with clear inputs, formula shown and shareable results.

About Grid Search Combinations Calculator

A grid search evaluates the full cross product of the parameter values, and cross-validation multiplies each combination by the fold count. Sixty combinations with five folds is 300 model fits, which is why the total fit count — not the combination count — is the number to check against your compute budget before starting.

Formula

Grid search cost

combinations = product of values per parameter; fits = combinations x folds

Worked example

With the default values below, Grid Search Combinations Calculator returns Model fits the search will run of 300. Change any input above and every figure here is recalculated in your browser.

Inputs used

Values for parameter 1minimum 1: 4
Values for parameter 2minimum 1: 5
Values for parameter 3minimum 1: 3
Cross-validation foldsminimum 1: 5

Results

Model fits the search will run: 300
Hyperparameter combinations: 60
Fits per combination: 5
Trials random search would need instead: 60

Step by step

1Values used
2Grid search cost
3Model fits the search will run= 300
4Hyperparameter combinations= 60
5Fits per combination= 5
6Trials random search would need instead= 60

Assumptions and limits

The result is derived only from the 4 inputs listed above. Anything not among them is not part of the calculation.
In particular, this calculator does not take inflation, tax, fees or charges or compounding frequency into account.
Figures are rounded for display to whole numbers, so totals built from several rounded parts can differ slightly from the unrounded arithmetic.

How these calculators are built and tested is described on the Methodology page. If a figure looks wrong, please report it.

Frequently Asked Questions

Does refitting on the full data add a fit?

Yes, most implementations refit the best configuration on all the data at the end, so add one to the total.

Can the search be shortened without losing much?

Halving strategies such as successive halving and HyperBand allocate few resources to most candidates and only train the promising ones fully, often cutting total cost by an order of magnitude.