Fitting Deep Distributional Regression

Documentation for package ‘deepregression’ version 1.0.0

Help Pages

-- C --

check_and_install	Function to check python environment and install necessary packages
coef.deepregression	Generic functions for deepregression models
coef.drEnsemble	Method for extracting ensemble coefficient estimates
combine_penalties	Function to combine two penalties
create_family	Function to create (custom) family
create_penalty	Function to create mgcv-type penalty
cv	Generic cv function
cv.deepregression	Generic functions for deepregression models

-- D --

deepregression	Fitting Semi-Structured Deep Distributional Regression
distfun_to_dist	Function to define output distribution based on dist_fun

-- E --

ensemble	Generic deep ensemble function
ensemble.deepregression	Ensemblind deepregression models
extractlen	Formula helpers
extractval	Formula helpers
extractvar	Extract variable from term
extract_pure_gam_part	Extract the smooth term from a deepregression term specification
extract_S	Convenience function to extract penalty matrix and value

-- F --

family_to_tfd	Character-tfd mapping function
family_to_trafo	Character-to-transformation mapping function
fit.deepregression	Generic functions for deepregression models
fitted.deepregression	Generic functions for deepregression models
fitted.drEnsemble	Method for extracting the fitted values of an ensemble
form2text	Formula helpers
form_control	Options for formula parsing
from_dist_to_loss	Function to transform a distritbution layer output into a loss function
from_preds_to_dist	Define Predictor of a Deep Distributional Regression Model

-- G --

gam_plot_data	used by gam_processor
gam_processor	Function that creates layer for each processor
get_distribution	Function to return the fitted distribution
get_ensemble_distribution	Obtain the conditional ensemble distribution
get_gamdata	Extract property of gamdata
get_gamdata_reduced_nr	Extract number in matching table of reduced gam term
get_gam_part	Extract gam part from wrapped term
get_layernr_by_opname	Function to return layer number given model and name
get_layernr_trainable	Function to return layer numbers with trainable weights
get_layer_by_opname	Function to return layer given model and name
get_names_pfc	Extract term names from the parsed formula content
get_partial_effect	Return partial effect of one smooth term
get_processor_name	Extract processor name from term
get_special	Extract terms defined by specials in formula
get_type_pfc	Function to subset parsed formulas
get_weight_by_name	Function to retrieve the weights of a structured layer
get_weight_by_opname	Function to return weight given model and name

-- H --

Function to define smoothness and call mgcv's smooth constructor

-- I --

int_processor	Function that creates layer for each processor
inverse_group_lasso_pen	Hadamard-type layers

-- K --

Compile a Deep Distributional Regression Model

-- L --

layer_add_identity	Convenience layer function
layer_concatenate_identity	Convenience layer function
layer_generator	Function that creates layer for each processor
layer_group_hadamard	Hadamard-type layers
layer_hadamard	Hadamard-type layers
layer_hadamard_diff	Hadamard-type layers
layer_sparse_conv_2d	Sparse 2D Convolutional layer
layer_spline	Function to define spline as TensorFlow layer
lin_processor	Function that creates layer for each processor
log_score	Function to return the log_score
loop_through_pfc_and_call_trafo	Function to loop through parsed formulas and apply data trafo

-- M --

makeInputs	Convenience layer function
makelayername	Function that takes term and create layer name
make_folds	Generate folds for CV out of one hot encoded matrix
make_generator	creates a generator for training
make_generator_from_matrix	Make a DataGenerator from a data.frame or matrix
make_tfd_dist	Families for deepregression
mean.deepregression	Generic functions for deepregression models
multioptimizer	Function to define an optimizer combining multiple optimizers

-- N --

Returns the parameter names for a given family

-- O --

orthog_control	Options for orthogonalization
orthog_P	Function to compute adjusted penalty when orthogonalizing
orthog_post_fitting	Orthogonalize a Semi-Structured Model Post-hoc
orthog_structured_smooths_Z	Orthogonalize structured term by another matrix

-- P --

penalty_control	Options for penalty setup in the pre-processing
plot.deepregression	Generic functions for deepregression models
plot_cv	Plot CV results from deepregression
precalc_gam	Pre-calculate all gam parts from the list of formulas
predict.deepregression	Generic functions for deepregression models
predict_gen	Generator function for deepregression objects
prepare_data	Function to prepare data based on parsed formulas
prepare_newdata	Function to prepare new data based on parsed formulas
print.deepregression	Generic functions for deepregression models
process_terms	Control function to define the processor for terms in the formula

-- Q --

quant	Generic quantile function
quant.deepregression	Generic functions for deepregression models

-- R --

regularizer_group_lasso	Hadamard-type layers
reinit_weights	Genereic function to re-intialize model weights
reinit_weights.deepregression	Method to re-initialize weights of a '"deepregression"' model

-- S --

separate_define_relation	Function to define orthogonalization connections in the formula
simplyconnected_layer	Hadamard-type layers
stddev	Generic sd function
stddev.deepregression	Generic functions for deepregression models
stop_iter_cv_result	Function to get the stoppting iteration from CV
subnetwork_init	Initializes a Subnetwork based on the Processed Additive Predictor

-- T --

tfd_mse	For using mean squared error via TFP
tfd_zinb	Implementation of a zero-inflated negbinom distribution for TFP
tfd_zip	Implementation of a zero-inflated poisson distribution for TFP
tf_repeat	TensorFlow repeat function which is not available for TF 2.0
tf_row_tensor	Row-wise tensor product using TensorFlow
tf_split_multiple	Split tensor in multiple parts
tf_stride_cols	Function to index tensors columns
tf_stride_last_dim_tensor	Function to index tensors last dimension
tibgroup_layer	Hadamard-type layers
tib_layer	Hadamard-type layers

-- U --

Function to update miniconda and packages

-- W --

Options for weights of layers