processResults: Process results: return imputation metrics

In DavidKLim/NIMIWAE: Non-Ignorably Missing Importance Weighted Autoencoder

Process results: return imputation metrics

Description

Process results: return imputation metrics

Usage

processResults(
  data = NULL,
  Missing = NULL,
  g = NULL,
  res = NULL,
  data.file.name = "",
  res.file.name = "",
  method = c("MIWAE", "NIMIWAE", "HIVAE", "VAEAC", "MEAN", "MF")
)

Arguments

data	Data matrix (N x P)
Missing	Missingness mask matrix (N x P)
g	Training-validation-test split partitioning
res	Results object output from NIMIWAE function
data.file.name	Path to data file, which contains "data", "Missing", and "g". These inputs need not be specified if data.file.name is specified
res.file.name	Path to res file, which contains "res_<method>" results object from the method that was run. "res" need not be specified if res.file.name is specified
method	Method used for imputation. "NIMIWAE" is used for this package, but results from competing methods ("MIWAE", "HIVAE", "VAEAC", "MEAN", "MF") can also be processed. See the NIMIWAE_Paper repo for more details

Value

list of objects: res (original res input), results (named vector of imputation metrics), and call (user-input call to function)

Author(s)

David K. Lim, [e-mail]

References

https://github.com/DavidKLim/NIMIWAE

Examples

processResults(data.file.name="Results/CONCRETE/data_MCAR_25.RData", res.file.name="Results/CONCRETE/res_NIMIWAE_MCAR_25_IWAE_rzF.RData", method="NIMIWAE")

DavidKLim/NIMIWAE documentation built on Jan. 19, 2024, 11:18 p.m.