deep_complier_mod: Train complier model using deep neural learning through...
In DeepLearningCausal: Causal Inference with Super Learner and Deep Neural Networks

deep_complier_mod

R Documentation

Train complier model using deep neural learning through Tensorflow

Description

Train model using group exposed to treatment with compliance as binary outcome variable and covariates.

Usage

deep_complier_mod(
  complier.formula,
  exp.data,
  treat.var,
  algorithm = "adam",
  hidden.layer = c(2, 2),
  hidden_activation = "relu",
  ID = NULL,
  epoch = 10,
  verbose = 1,
  batch_size = 32,
  validation_split = NULL,
  patience = NULL,
  dropout_rate = NULL
)

Arguments

`complier.formula`	formula to fit compliance model (c ~ x) using complier variable and covariates
`exp.data`	list object of experimental data.
`treat.var`	string specifying the binary treatment variable
`algorithm`	string for name of optimizer algorithm. Set to adam. other optimization algorithms available are sgd, rprop, adagrad.
`hidden.layer`	vector specifying the hidden layers and the number of neurons in each layer.
`hidden_activation`	string or vector for activation function used for hidden layers. Defaults to "relu".
`ID`	string for name of identifier variable.
`epoch`	integer for number of epochs
`verbose`	1 to display model training information and learning curve plot. 0 to suppress messages and plots.
`batch_size`	integer for batch size to split the training set. Defaults to 32.
`validation_split`	double for proportion of training data to be split for validation.
`patience`	integer for number of epochs with no improvement after which training will be stopped.
`dropout_rate`	double or vector for proportion of hidden layer to drop out.