Man pages for tsensembler
Dynamic Ensembles for Time Series Forecasting

ADEArbitrated Dynamic Ensemble
ADE-classArbitrated Dynamic Ensemble
ade_hatPredictions by an ADE ensemble
ade_hat-classPredictions by an ADE ensemble
aeComputing the absolute error
base_ensemblebase_ensemble
base_ensemble-classbase_ensemble-class
base_models_lossComputing the error of base models
best_mvrGet best PLS/PCR model
blocked_prequentialPrequential Procedure in Blocks
bm_cubistFit Cubist models (M5)
bm_ffnnFit Feedforward Neural Networks models
bm_gaussianprocessFit Gaussian Process models
bm_gbmFit Generalized Boosted Regression models
bm_glmFit Generalized Linear Models
bm_marsFit Multivariate Adaptive Regression Splines models
bm_pls_pcrFit PLS/PCR regression models
bm_pprFit Projection Pursuit Regression models
bm_randomforestFit Random Forest models
bm_svrFit Support Vector Regression models
build_base_ensembleWrapper for creating an ensemble
build_committeeBuilding a committee for an ADE model
build_committee_setBuild committee set
CA.ADE_hatCA generaliser using arbitrage
CA.EWA_hatCA generaliser using exponentially weighted average
CA.FixedShare_hatCA generaliser using fixed share
CA.MLpol_hatCA generaliser using polynomial weighted average
CA.OGD_hatCA generaliser using OGD
CA.Ridge_hatCA generaliser using ridge regression
combine_committeesMerge across sub-ensembles
combine_predictionsCombining the predictions of several models
committee_set-classcommittee_set-class
compute_predictionsCompute the predictions of base models
constructive_aggregationConstructive aggregation constructor
constructive_aggregation_constructive_aggregation_
constructive_aggregation-classconstructive_aggregation-class
contiguous_countContiguity check
DETSDynamic Ensemble for Time Series
DETS-classDynamic Ensemble for Time Series
dets_hatPredictions by an DETS ensemble
dets_hat-classPredictions by an DETS ensemble
EMASEWeighting Base Models by their Moving Average Squared Error
embed_timeseriesEmbedding a Time Series
erfcComplementary Gaussian Error Function
FIFOFirst-In First Out
forecastForecasting using an ensemble predictive model
get_targetGet the target from a formula
get_top_modelsExtract top learners from their weights
get_yGet the response values from a data matrix
hat_infoGet predict data for generalising
holdoutHoldout
intraining_estimationsOut-of-bag loss estimations
intraining_predictionsOut-of-bag predictions
l1applyApplying lapply on the rows
learning_base_modelsTraining the base models of an ensemble
loss_meta_learnTraining an arbiter
maeComputing the mean absolute error
maseComputing the mean absolute scaled error
merging_in_expertsMerge models in each committee
meta_gpTraining a Gaussian process arbiter
meta_gp_predictArbiter predictions via linear model
meta_lassoTraining a LASSO arbiter
meta_lasso_predictArbiter predictions via linear model
meta_predictPredicting loss using arbiter
meta_rfTraining a random forest arbiter
meta_rf_predictArbiter predictions via ranger
model_recent_performanceRecent performance of models using EMASE
model_specsSetup base learning models
model_specs-classSetup base learning models
model_weightingModel weighting
mseComputing the mean squared error
normalizeScale a numeric vector using max-min
predict-constructive_aggregation-methodpredict method for constructive aggregation
predict-methodsPredicting new observations using an ensemble
predict_pls_pcrpredict method for pls/pcr
proportionComputing the proportions of a numeric vector
prune_c_contiguityPrune subsets by contiguity
prune_c_outperformancePrune subsets by out-performance
rbind_lrbind with do.call syntax
recent_lambda_observationsGet most recent lambda observations
rm.nullList without null elements
rmseComputing the root mean squared error
roll_mean_matrixComputing the rolling mean of the columns of a matrix
r_squaredComputing R squared
seComputing the squared error
select_bestSelecting best model according to weights
sequential_reweightingSequential Re-weighting for controlling predictions'...
sliding_similaritySliding similarity via Pearson's correlation
soft.completionSoft Imputation
softmaxComputing the softmax
split_bySplitting expressions by pattern
train_adeTraining procedure of for ADE
tsensemblerDynamic Ensembles for Time Series Forecasting
unlistnUnlist not using names
update_adeUpdating an ADE model
update_ade_metaUpdating the metalearning layer of an ADE model
update_base_modelsUpdate the base models of an ensemble
update_weightsUpdating the weights of base models
vcapplyvapply extension for character values
viapplyvapply extension for integer values
vlapplyvapply extension for logical values
vnapplyvapply extension for numeric values
water_consumptionWater Consumption in Oporto city (Portugal) area.
tsensembler documentation built on April 14, 2018, 1:03 a.m.