diss_to_neighbors

a matrix representing the dissimilarities between
observations in a matrix <code>Xu</code> and observations in another matrix
<code>Xr</code>. <code>Xr</code> in rows <code>Xu</code> in columns.

diss_matrix

an integer value indicating the k-nearest neighbors of each
observation in <code>Xu</code> that must be selected from <code>Xr</code>.

an integer value indicating a dissimilarity treshold.
For each observation in <code>Xu</code>, its nearest neighbors in <code>Xr</code>
are selected as those for which their dissimilarity to <code>Xu</code> is below
this <code>k_diss</code> threshold. This treshold depends on the corresponding
dissimilarity metric specified in <code>diss_method</code>. Either <code>k</code> or
<code>k_diss</code> must be specified.

k_diss

an integer vector of length 2 which specifies the minimum
(first value) and the maximum (second value) number of neighbors to be
retained when the <code>k_diss</code> is given.

k_range

a vector of integers indicating what observations in <code>Xr</code>
(and <code>Yr</code>) must be 'forced' to always be part of all the neighborhoods.

spike

logical indicating if the input dissimilarity
must be mirroed in the output.

return_dissimilarity

internal

Functions for dissimilarity analysis and memory-based learning
(MBL, a.k.a local modeling) in complex spectral data sets.
Most of these functions are based on the methods presented in
Ramirez-Lopez et al. (2013) <doi:10.1016/j.geoderma.2012.12.014>.

Leonardo RamirezLopez

resemble

Memory-Based Learning in Spectral Chemometrics

diss_to_neighbors function

<dl><dt>diss_matrix</dt>
<dd>a matrix representing the dissimilarities between
observations in a matrix <code>Xu</code> and observations in another matrix
<code>Xr</code>. <code>Xr</code> in rows <code>Xu</code> in columns.</dd>
<dt>k</dt>
<dd>an integer value indicating the k-nearest neighbors of each
observation in <code>Xu</code> that must be selected from <code>Xr</code>.</dd>
<dt>k_diss</dt>
<dd>an integer value indicating a dissimilarity treshold.
For each observation in <code>Xu</code>, its nearest neighbors in <code>Xr</code>
are selected as those for which their dissimilarity to <code>Xu</code> is below
this <code>k_diss</code> threshold. This treshold depends on the corresponding
dissimilarity metric specified in <code>diss_method</code>. Either <code>k</code> or
<code>k_diss</code> must be specified.</dd>
<dt>k_range</dt>
<dd>an integer vector of length 2 which specifies the minimum
(first value) and the maximum (second value) number of neighbors to be
retained when the <code>k_diss</code> is given.</dd>
<dt>spike</dt>
<dd>a vector of integers indicating what observations in <code>Xr</code>
(and <code>Yr</code>) must be 'forced' to always be part of all the neighborhoods.</dd>
<dt>return_dissimilarity</dt>
<dd>logical indicating if the input dissimilarity
must be mirroed in the output.</dd></dl>

Arguments

From dissimilarity matrix to neighbors — diss_to_neighbors

<dl>

<dt>diss_matrix</dt>
<dd>a matrix representing the dissimilarities between
observations in a matrix <code>Xu</code> and observations in another matrix
<code>Xr</code>. <code>Xr</code> in rows <code>Xu</code> in columns.</dd>


<dt>k</dt>
<dd>an integer value indicating the k-nearest neighbors of each
observation in <code>Xu</code> that must be selected from <code>Xr</code>.</dd>


<dt>k_diss</dt>
<dd>an integer value indicating a dissimilarity treshold.
For each observation in <code>Xu</code>, its nearest neighbors in <code>Xr</code>
are selected as those for which their dissimilarity to <code>Xu</code> is below
this <code>k_diss</code> threshold. This treshold depends on the corresponding
dissimilarity metric specified in <code>diss_method</code>. Either <code>k</code> or
<code>k_diss</code> must be specified.</dd>


<dt>k_range</dt>
<dd>an integer vector of length 2 which specifies the minimum
(first value) and the maximum (second value) number of neighbors to be
retained when the <code>k_diss</code> is given.</dd>


<dt>spike</dt>
<dd>a vector of integers indicating what observations in <code>Xr</code>
(and <code>Yr</code>) must be 'forced' to always be part of all the neighborhoods.</dd>


<dt>return_dissimilarity</dt>
<dd>logical indicating if the input dissimilarity
must be mirroed in the output.</dd>

</dl>

diss_to_neighbors: From dissimilarity matrix to neighbors

Description

Usage

Arguments