mbl_control: A function that controls some few aspects of the memory-based...

Description Usage Arguments Details Value Author(s) References See Also Examples

View source: R/mbl_control.R

Description

\loadmathjax

This function is used to further control some aspects of the memory-based learning process in the mbl function.

Usage

1
2
3
4
5
6
7
mbl_control(return_dissimilarity = FALSE,
            validation_type = c("NNv", "local_cv"),
            tune_locally = TRUE,
            number = 10,
            p = 0.75,
            range_prediction_limits = TRUE,
            allow_parallel = TRUE)

Arguments

return_dissimilarity

a logical indicating if the dissimilarity matrix between Xr and Xu must be returned.

validation_type

a character vector which indicates the (internal) validation method(s) to be used for assessing the global performance of the local models. Possible options are: "NNv" and "local_cv". Alternatively "none" can be used when cross-validation is not required (see details below).

tune_locally

a logical. It only applies when validation_type = "local_cv" and "pls" or "wapls" fitting algorithms are used. If TRUE, then the parameters of the local pls-based models (i.e. pls factors for the "pls" method and minimum and maximum pls factors for the "wapls" method) are optimized. Default is TRUE.

number

an integer indicating the number of sampling iterations at each local segment when "local_cv" is selected in the validation_type argument. Default is 10.

p

a numeric value indicating the percentage of calibration observations to be retained at each sampling iteration at each local segment when "local_cv" is selected in the validation_type argument. Default is 0.75 (i.e. 75 "%").

range_prediction_limits

a logical. It indicates whether the prediction limits at each local regression are determined by the range of the response variable within each neighborhood. When the predicted value is outside this range, it will be automatically replaced with the value of the nearest range value. If FALSE, no prediction limits are imposed. Default is TRUE.

allow_parallel

a logical indicating if parallel execution is allowed. If TRUE, this parallelism is applied to the loop in mbl in which each iteration takes care of a single observation in Xu. The parallelization of this for loop is implemented using the codeforeach function of the package foreach. Default is TRUE.

Details

The validation methods available for assessing the predictive performance of the memory-based learning method used are described as follows:

Value

a list mirroring the specified parameters

Author(s)

Leonardo Ramirez-Lopez and Antoine Stevens

References

Ramirez-Lopez, L., Behrens, T., Schmidt, K., Stevens, A., Dematte, J.A.M., Scholten, T. 2013a. The spectrum-based learner: A new local approach for modeling soil vis-NIR spectra of complex data sets. Geoderma 195-196, 268-279.

Ramirez-Lopez, L., Behrens, T., Schmidt, K., Viscarra Rossel, R., Dematte, J. A. M., Scholten, T. 2013b. Distance and similarity-search metrics for use with soil vis-NIR spectra. Geoderma 199, 43-53.

See Also

f_diss, cor_diss, sid, ortho_diss, mbl

Examples

1
2
# A control list with the default parameters
mbl_control()

resemble documentation built on Nov. 9, 2020, 5:08 p.m.