Skip to contents

Calculates predicted values from a fitted Hmsc model

Source: R/Mod_Predict_Hmsc.R
Predict_Hmsc.Rd

This function modifies the Hmsc:::predict.Hmsc function.

Usage

Predict_Hmsc(
  Path_Model,
  Loff = NULL,
  XData = NULL,
  X = NULL,
  XRRRData = NULL,
  XRRR = NULL,
  Gradient = NULL,
  Yc = NULL,
  mcmcStep = 1L,
  expected = TRUE,
  NCores = 8L,
  Model_Name = "Train",
  Temp_Dir = "TEMP_Pred",
  Temp_Cleanup = TRUE,
  RC = NULL,
  UseTF = TRUE,
  TF_Environ = NULL,
  TF_use_single = FALSE,
  LF_OutFile = NULL,
  LF_Return = FALSE,
  LF_InputFile = NULL,
  LF_Only = FALSE,
  LF_NCores = NCores,
  LF_Check = FALSE,
  LF_Temp_Cleanup = TRUE,
  LF_Commands_Only = FALSE,
  Pred_Dir = NULL,
  Pred_PA = NULL,
  Pred_XY = NULL,
  Evaluate = FALSE,
  Eval_Name = NULL,
  Eval_Dir = "Evaluation",
  Verbose = TRUE
)

Arguments

Path_Model

Character. Path to the saved model object.

Loff

See Hmsc::predict.Hmsc for more details.

XData

data.frame. The unpreprocessed covariates for the predictions to be made. Works only if the XFormula argument was specified in the Hmsc::Hmsc model constructor call. Requirements are similar to those in the Hmsc model constructor.

X

matrix. Covariates for the predictions to be made. Only one of XData and X arguments may be provided.

XRRRData

data.frame. Covariates for reduced-rank regression.

XRRR

matrix. Covariates for reduced-rank regression.

Gradient

An object returned by Hmsc::constructGradient. Providing Gradient is an alternative for providing XData, studyDesign and ranLevels. Cannot be used together with Yc.

Yc

matrix. Outcomes that are assumed to be known for conditional predictions. Cannot be used together with Gradient.

mcmcStep

Integer. Number of extra mcmc steps used for updating the random effects. Defaults to 1L.

expected

Logical. Whether to return the location parameter of the observation models or sample the values from those. Defaults to TRUE.

NCores

Integer. Number of CPU cores to use for parallel processing. Default: 8.

Model_Name

Character. Prefix for temporary file names. Defaults to NULL, in which case no prefix is used.

Temp_Dir

Character. Path for temporary storage of intermediate files.

Temp_Cleanup

Logical. Whether to clean up temporary files. Defaults to TRUE.

RC

Character. Type of predictions to be made. If NULL (default), predictions are made for the latent factors. If c, predictions are made for response curves at mean coordinates. If i, predictions are made for response curves at infinite coordinates.

UseTF

Logical. Whether to use TensorFlow for calculations. Defaults to TRUE.

TF_Environ

Character. Path to the Python environment. This argument is required if UseTF is TRUE under Windows. Defaults to NULL.

TF_use_single

Logical. Whether to use single precision for the TensorFlow calculations. Defaults to FALSE.

LF_OutFile

Character. Path to save the outputs. If NULL (default), the predicted latent factors are not saved to a file. This should end with either *.qs2 or *.RData.

LF_Return

Logical. Whether the output should be returned. Defaults to FALSE. If LF_OutFile is NULL, this parameter cannot be set to FALSE because the function needs to return the result if it is not saved to a file.

LF_InputFile

Character. File name where the latent factor predictions are saved. If NULL (default), latent factor predictions will be computed. If specified, latent factor predictions are read from this path. This allows to predicting the latent factors for new sites only once.

LF_Only

Logical. Whether to return the latent factor predictions only. Defaults to FALSE. This helps in predicting to new sites, allowing to predicting the latent factors only once, then the output can be loaded in other predictions when needed.

LF_NCores

Integer. Number of cores to use for parallel processing of latent factor prediction. Defaults to 8L.

LF_Check

Logical. If TRUE, the function checks if the output files are already created and valid. If FALSE, the function will only check if the files exist without checking their integrity. Default is FALSE.

LF_Temp_Cleanup

Logical. Whether to delete temporary files in the Temp_Dir directory after finishing the LF predictions.

LF_Commands_Only

Logical. If TRUE, returns the command to run the Python script. Default is FALSE.

Pred_Dir

Character. Directory path indicating where the predictions will be saved. Defaults to NULL, which saves model predictions to "Model_Prediction" folder of the current working directory.

Pred_PA

matrix. Presence-absence data for evaluation. If NULL (default), the presence-absence data from the model object is used. This argument is used only when Evaluate is TRUE.

Pred_XY

matrix. Coordinates to be added to predicted values. If NULL (default), the coordinates from the model object is used.

Evaluate

Logical. Whether to evaluate the model predictions. Defaults to FALSE.

Eval_Name

Character. Name of the evaluation results. If NULL, the default name is used (Eval_[Model_Name].qs2).

Eval_Dir

Character. Directory where the evaluation results will be saved. Defaults to Evaluation.

Verbose

Logical. Whether to print a message upon successful saving of files. Defaults to FALSE.