Hyperparameter (machine learning)
In the context of machine learning, grid search is an algorithm that performs an exhaustive search through the parameter space of a learning algorithm to solve problem of model selection: finding the optimal parameters for a dataset. The algorithm must be giuded by some performance metric, measured by cross validation on a training set.[1]
External link
- The machine learning toolkit scikit-learn includes a grid search module.
References
- ^ Chin-Wei Hsu, Chih-Chung Chang and Chih-Jen Lin (2010). A practical guide to support vector classification. Technical Report, National Taiwan University.