from_preds_to_dist: Define Predictor of a Deep Distributional Regression Model

Description

Define Predictor of a Deep Distributional Regression Model

Usage

from_preds_to_dist(
  list_pred_param,
  family = NULL,
  output_dim = 1L,
  mapping = NULL,
  from_family_to_distfun = make_tfd_dist,
  from_distfun_to_dist = distfun_to_dist,
  add_layer_shared_pred = function(x, units) layer_dense(x, units = units, use_bias =
    FALSE),
  trafo_list = NULL
)

Value

a list with input tensors and output tensors that can be passed to, e.g., keras_model

Arguments

list_pred_param: list of input-output(-lists) generated from subnetwork_init
family: see ?deepregression; if NULL, concatenated list_pred_param entries are returned (after applying mapping if provided)
output_dim: dimension of the output
mapping: a list of integers. The i-th list item defines which element elements of list_pred_param are used for the i-th parameter. For example, mapping = list(1,2,1:2) means that list_pred_param[[1]] is used for the first distribution parameter, list_pred_param[[2]] for the second distribution parameter and list_pred_param[[3]] for both distribution parameters (and then added once to list_pred_param[[1]] and once to list_pred_param[[2]])
from_family_to_distfun: function to create a dist_fun (see ?distfun_to_dist) from the given character family
from_distfun_to_dist: function creating a tfp distribution based on the prediction tensors and dist_fun. See ?distfun_to_dist
add_layer_shared_pred: layer to extend shared layers defined in mapping
trafo_list: a list of transformation function to convert the scale of the additive predictors to the respective distribution parameter