Package index • GreedyExperimentalDesign

Package overview

GreedyExperimentalDesign: Greedy Experimental Design Search

Randomization and blocking

complete_randomization(): Implements complete randomization (without forced balance)
complete_randomization_with_forced_balanced(): Implements forced balanced randomization
imbalanced_block_designs(): Implements unequally allocated block designs
imbalanced_complete_randomization(): Implements unequally allocated complete randomization
hadamardExperimentalDesign(): Create a Hadamard Design
generate_block_design_cpp_wrap(): Generates homogeneous block design allocations rapidly
gen_var_cov_matrix_block_designs(): Computes varcov matrix for block designs
computeBinaryMatchStructure(): Compute Binary Matching Strcuture
optimize_asymmetric_treatment_assignment(): Compute Optimal Number of Treatments/Controls

Design searches

initBinaryMatchExperimentalDesignSearchObject(): Begin a Binary Match Search
initBinaryMatchFollowedByGreedyExperimentalDesignSearchObject(): Begin a Search for Binary Matching Followed by Greedy Switch Designs
initBinaryMatchFollowedByRerandomizationDesignSearchObject(): Begin a Search for Binary Matching Followed by Rerandomization
initGreedyExperimentalDesignObject(): Begin A Greedy Pair Switching Search
initGreedyMultipleKernelExperimentalDesignObject(): Begin A Greedy Multiple Kernel Design Search
initGurobiNumericalOptimizationExperimentalDesignObject(): Begin Gurobi Optimized Search
initKarpExperimentalDesignObject(): Begin Karp Search
initOptimalExperimentalDesignObject(): Begin a Search for the Optimal Solution
initRerandomizationExperimentalDesignObject(): Begin a Rerandomization Search
startSearch(): Starts the parallelized greedy design search.
stopSearch(): Stops the parallelized greedy design search.
searchTimeElapsed(): Returns the amount of time elapsed

Search results and summaries

resultsBinaryMatchSearch(): Binary Pair Match Search
resultsBinaryMatchThenGreedySearch(): Returns unique allocation vectors that are binary matched
resultsBinaryMatchThenRerandomizationSearch(): Returns unique allocation vectors that are binary matched
resultsGreedySearch(): Returns the results (thus far) of the greedy design search
resultsGurobiNumericalOptimizeSearch(): Query the Gurobi Results
resultsKarpSearch(): Returns the results (thus far) of the karp design search
resultsMultipleKernelGreedySearch(): Returns the results of a greedy multiple kernel search
resultsOptimalSearch(): Returns the results (thus far) of the optimal design search
resultsRerandomizationSearch(): Returns the results (thus far) of the rerandomization design search
summary(<binary_match_structure>): Prints a summary of a binary_match_structure object
summary(<binary_then_greedy_experimental_design>): Prints a summary of a binary_then_greedy_experimental_design object
summary(<binary_then_rerandomization_experimental_design>): Prints a summary of a binary_then_rerandomization_experimental_design object
summary(<greedy_experimental_design_search>): Prints a summary of a greedy_experimental_design_search object
summary(<greedy_multiple_kernel_experimental_design>): Prints a summary of a greedy_multiple_kernel_experimental_design object
summary(<karp_experimental_design_search>): Prints a summary of a karp_experimental_design_search object
summary(<optimal_experimental_design_search>): Prints a summary of a optimal_experimental_design_search object
summary(<pairwise_matching_experimental_design_search>): Prints a summary of a pairwise_matching_experimental_design_search object
summary(<rerandomization_experimental_design_search>): Prints a summary of a rerandomization_experimental_design_search object
print(<binary_match_structure>): Prints a summary of a binary_match_structure object
print(<binary_then_greedy_experimental_design>): Prints a summary of a binary_then_greedy_experimental_design object
print(<binary_then_rerandomization_experimental_design>): Prints a summary of a binary_then_rerandomization_experimental_design object
print(<greedy_experimental_design_search>): Prints a summary of a greedy_experimental_design_search object
print(<greedy_multiple_kernel_experimental_design>): Prints a summary of a greedy_multiple_kernel_experimental_design object
print(<karp_experimental_design_search>): Prints a summary of a karp_experimental_design_search object
print(<optimal_experimental_design_search>): Prints a summary of a optimal_experimental_design_search object
print(<pairwise_matching_experimental_design_search>): Prints a summary of a pairwise_matching_experimental_design_search object
print(<rerandomization_experimental_design_search>): Prints a summary of a rerandomization_experimental_design_search object

GPU and numerical backends

ged_gpu_available(): Check for optional GPU support
ged_gpu_devices(): List optional GPU devices
compute_distance_matrix_gpu(): Compute a squared Euclidean distance matrix through the optional native backend
compute_kernel_matrix_gpu(): Compute a kernel matrix through the optional native backend
compute_multiple_kernel_objective_vals_gpu(): Compute multiple kernel objective values using GPU
compute_objective_vals_gpu(): Compute kernel objective values through the optional native backend
compute_randomization_metrics_gpu(): Compute randomization metrics through the optional native backend
full_greedy_search_gpu(): Run a full greedy search on GPU (Upload Once, In-Place)
compute_gram_matrix(): Gram Matrix Computation
compute_objective_val(): Computes Objective Value From Allocation Vector
safe_cov_inverse(): Computes a numerically stable inverse of a covariance matrix
standardize_data_matrix(): Standardizes the columns of a data matrix.

Plots and utilities

plot(<greedy_experimental_design_search>): Plots a summary of a greedy search object object
plot(<greedy_multiple_kernel_experimental_design>): Plots a summary of a greedy_multiple_kernel_experimental_design object
plot_obj_val_by_iter(): Plots the objective value by iteration
plot_obj_val_order_statistic(): Plots an order statistic of the object value as a function of number of searches
compute_distance_matrix_cpp_wrap(): Computes a Euclidean-squared distance matrix rapidly
compute_distance_matrix_gpu(): Compute a squared Euclidean distance matrix through the optional native backend
compute_gram_matrix(): Gram Matrix Computation
compute_kernel_matrix_gpu(): Compute a kernel matrix through the optional native backend
compute_multiple_kernel_objective_vals_gpu(): Compute multiple kernel objective values using GPU
compute_objective_val(): Computes Objective Value From Allocation Vector
compute_objective_vals_gpu(): Compute kernel objective values through the optional native backend
compute_randomization_metrics(): Computes Randomization Metrics (explained in paper) about a design algorithm
compute_randomization_metrics_gpu(): Compute randomization metrics through the optional native backend
gen_pm_designs_cpp_wrap(): Create PM designs
gen_var_cov_matrix_block_designs(): Computes varcov matrix for block designs
create_all_ys_cpp_wrap(): Create all binary Y's convenience function using a randomized design
shuffle_cpp_wrap(): Shuffles a vector rapidly
all_elements_same_cpp_wrap(): Tests if a vector has all elements the same
generate_stdzied_design_matrix(): Generates a design matrix with standardized predictors.
greedy_orthogonalization_curation(): Curate More Orthogonal Vectors Greedily
greedy_orthogonalization_curation2(): Curate More Orthogonal Vectors Greedily