Grid search
Grid search is a search method that evaluates configurations formed from specified candidate values, typically every combination in their Cartesian product.
Grokking Machine Learning
A model-selection technique that evaluates an exhaustive set of specified hyperparameter combinations.
- How it is used
- Selects hyperparameters by comparing models trained with different combinations on a validation set.
- How deep it goes
- Defines the search and demonstrates candidate combinations, validation scoring, and Scikit-learn implementation.
- Examples
- Uses an SVM with an RBF kernel, varying gamma and C and comparing validation accuracy.
- What you can do after
- Apply it via Scikit-learn's GridSearchCV to search parameter combinations and retrieve the best estimator.
Evolutionary Deep Learning: Genetic algorithms and neural networks
Grid search is a hyperparameter-optimization technique that evaluates configurations formed from a specified Cartesian product of candidate values.
- How it is used
- Automates hyperparameter optimization as a systematic baseline before applying more advanced search methods.
- How deep it goes
- Explains grid construction, iteration, visualization, and resolution trade-offs, but stops short of evolutionary optimization.
- Examples
- Uses Python generators, scikit-learn ParameterGrid, a 90-cell neural-network configuration, and plotted search results.
- What you can do after
- Apply grid search via Python and scikit-learn ParameterGrid to evaluate discrete hyperparameter combinations and inspect promising regions.
Read how Evolutionary Deep Learning: Genetic algorithms and neural networks explains grid search →
Designing Deep Learning Systems: A guide for software engineers
A hyperparameter search method that evaluates configurations formed from the Cartesian product of specified candidate values.
- How it is used
- Deploys grid search as a model-free method for configuring and launching hyperparameter-optimization trials.
- How deep it goes
- Defines Cartesian-product construction and operational drawbacks, but defers mathematical reasoning and implementation details.
- Examples
- Uses learning-rate and batch-size candidate sets to construct configurations such as (0.1, 10).
- What you can do after
- Operate/configure grid search by defining candidate values, generating their Cartesian product, and launching trials for each combination.
Read how Designing Deep Learning Systems: A guide for software engineers explains grid search →
Bayesian Optimization in Action
Grid search is an optimization technique that evaluates points arranged on a regular grid over the search space.
- How it is used
- Uses evenly spaced evaluations as a baseline for exploring an objective-function search space.
- How deep it goes
- Defines segment partitioning and endpoint evaluation, but stops short of adaptive narrowing or reliable global-optimum discovery.
- Examples
- Evaluates evenly spaced points from −5 to 5 and selects the point at 5 with an objective value near 4.86.
- What you can do after
- Apply it by partitioning a search space evenly and evaluating each segment endpoint.
Read how Bayesian Optimization in Action explains grid search →
Ensemble Methods for Machine Learning
Grid search is a model-selection method that evaluates combinations from a specified set of parameter values to identify a performant configuration.
- How it is used
- Selects interacting model hyperparameters by comparing cross-validated performance across a specified parameter grid.
- How deep it goes
- Explains parameter grids, scoring, folds, refitting, result inspection, and computational limits, but not search-theory internals.
- Examples
- Demonstrates AdaBoost digit classification with learning_rate and n_estimators grids, classification reports, and confusion matrices.
- What you can do after
- Apply it via scikit-learn’s GridSearchCV to select, refit, inspect, and test a tuned estimator.
Read how Ensemble Methods for Machine Learning explains grid search →
Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI
Grid search is an exhaustive hyperparameter optimization technique that evaluates every combination of specified hyperparameter values.
- How it is used
- Tunes model hyperparameters systematically after experimentation, especially when a small discrete search space is practical.
- How deep it goes
- Explains exhaustive combinations, search-grid structure, pipeline naming, validation, and computational limits without detailing the full implementation.
- Examples
- Uses logistic regression, regularization choices, C values, l1_ratio values, K-fold validation, and a reported score.
- What you can do after
- Apply it via Scikit-learn's GridSearchCV to select hyperparameters using cross-validated scores.
Read how Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI explains grid search →
Related concepts
- hyperparameteruses · A hyperparameter is a setting chosen outside ordinary parameter learning that controls a model, algorithm, or learning procedure.
- cross-validationuses · Cross-validation is a resampling technique that evaluates a model by repeatedly fitting it on some portions of available data and assessing it on held-out portions.
- random searchcontrasts with · Random search is an optimization technique that explores a search space by evaluating candidate solutions or configurations selected through randomized choices.
- Bayesian optimizationcontrasts with · Bayesian optimization is a sequential method for optimizing an expensive or unknown objective by using a probabilistic model to guide evaluations.
- linear regressionused by · Linear regression is a statistical modeling method that estimates a continuous response as a linear combination of predictor variables and fitted coefficients.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.