ParallelDSM (version 0.3.7)

ParallelComputing: ParallelComputing Functions

Description

ParallelComputing Functions

Usage

ParallelComputing(outpath, mymodels)

Arguments

outpath

: Output path of the result of the prediction file. The default is "output".

mymodels

: The models were selected, including QRF,RF and MLR.

Details

This function is the main function that performs parallel computations The outpath field refers to the filename of the data output The mymodels field has three modes to choose from: QRF,RF and MLR ‘QRF??? stands for Quantile Regression Forest Model Prediction Method ‘RF??? stands for Random Forest Model Prediction Method ‘MLR??? stands for Multiple Linear Regression Prediction Model

References

Breiman, L. (2001). Random forests. Mach. Learn. 45, 5???32. Meinshausen, N. (2006) "Quantile Regression Forests", Journal of Machine Learning Research 7, 983-999 http://jmlr.csail.mit.edu/papers/v7/

Examples

# \donttest{
## This function performs parallel computing, of which the parameters are as follows:
## outpath: the filename of the data output
## mymodels: which model user want to use. Three modes are available:
## Quantile Regression Forest (QRF),Random Forest (RF) and Multiple Linear Regression (MLR)

####################################################################################
# Example 1: Using random forest to produce soil map based on data in this package
# Loads related data sets
data("df.input" , package = "ParallelDSM")
data("df.mrrtf" , package = "ParallelDSM")
data("df.dem" , package = "ParallelDSM")

# Sets the path to the folder where the dataset will be stored
sampledata <- system.file("extdata" , "covariate", package = "ParallelDSM")

# Initializing the parameters for parallel computing
# ParallelInit_Test is same as ParallelInit
ParallelInit_Test(sampledata,df.input,dsmformul="socd030 ~ dem + mrrtf")
NormalizeData()
ParallelComputing(outpath = "mlrOutput" , mymodels = "MLR")
###################################################################################


###################################################################################
##  Example 2: Performing soil mapping based on my data with 3 CPUs ##

myinput <- "./all.input.csv"
# The sample data represents the file name where the data file is stored

# 'covariate' is the path name of a file
sampledata <- "./covariate" # the directory and filename
# The third parameter represents the name of the TIF file.
# nblock is used to partition the tif data into several blocks in the terms of row
# An appropriate nblock may optimize the speedup of parallel computing
ParallelInit(myinput,sampledata,"socd030 ~ twi + dem", nblock = 30 , ncore = 3)

ParallelComputing(outpath = "qrfOutput" , mymodels = "QRF")
###################################################################################


# }