Compositional Maximum Likelihood Estimation

[Up] [Top]

Documentation for package ‘compositional.mle’ version 1.0.2

Help Pages

%>>% Sequential Solver Composition
%|% Parallel Solver Racing (Operator)
bfgs BFGS Solver
chain Chain Solvers with Early Stopping
clear_cache Clear derivative cache
compose Compose Multiple Solvers Sequentially
compose_transforms Compose Multiple Function Transformations
coordinate_ascent Coordinate Ascent Solver
fisher_scoring Fisher Scoring Solver
get_fisher Get Fisher information function from problem
get_score Get score function from problem
gradient_ascent Gradient Ascent Solver
grid_search Grid Search Solver
is_converged Check if solver converged
is_mle_constraint Check if object is an mle_constraint
is_mle_numerical Check if object is an mle_numerical
is_mle_problem Check if object is an mle_problem
is_tracing Check if tracing is enabled
lbfgsb L-BFGS-B Solver (Box Constrained)
mle_constraint Create domain constraint specification
mle_problem Create an MLE Problem Specification
mle_trace Create a Trace Configuration
nelder_mead Nelder-Mead Solver (Derivative-Free)
newton_raphson Newton-Raphson Solver
normal_sampler Normal Sampler Factory
num_iterations Get number of iterations
optimization_path Extract Optimization Path as Data Frame
penalty_elastic_net Elastic net penalty (combination of L1 and L2)
penalty_l1 L1 penalty function (LASSO)
penalty_l2 L2 penalty function (Ridge)
plot.mle_numerical Plot Optimization Convergence
plot.mle_trace_data Plot Trace Data Directly
print.mle_problem Create an MLE Problem Specification
print.mle_trace Create a Trace Configuration
print.mle_trace_data Print MLE Trace Data
race Race Multiple Solvers
race_operator Parallel Solver Racing (Operator)
random_search Random Search Solver
sim_anneal Simulated Annealing Solver
uniform_sampler Uniform Sampler Factory
unless_converged Conditional Refinement
update.mle_problem Update an mle_problem
with_penalty Add penalty term to log-likelihood
with_restarts Multiple Random Restarts
with_subsampling Create stochastic log-likelihood with subsampling