bisquares2D

Sets up a top-layer set of bisquare basis functions on a square 2D domain of
size 1 x 1 for modelling the process \(Y\). It returns a list of length 1 containing the basis functions and encapsulated
functions that evaluate the bisquare functions over inputs of different types. See Value for more details.

Deep compositional spatial models are standard spatial covariance
models coupled with an injective warping function of the spatial
domain. The warping function is constructed through a composition
of multiple elemental injective functions in a deep-learning
framework. The package implements two cases for the univariate setting; first,
when these warping functions are known up to some weights that
need to be estimated, and, second, when the weights in each layer are random.
In the multivariate setting only the former case is available.
Estimation and inference is done using `tensorflow`, which makes use of
graphics processing units.
For more details see Zammit-Mangion et al. (2022) <doi:10.1080/01621459.2021.1887741>,
Vu et al. (2022) <doi:10.5705/ss.202020.0156>,
Vu et al. (2023) <doi:10.1016/j.spasta.2023.100742>, and
Shao et al. (2025) <doi:10.48550/arXiv.2505.12548>.

Quan Vu

deepspat

Deep Compositional Spatial Models

Andrew Zammit-Mangion

Xuanjie Shao

bisquares2D function

<dl><dt>r</dt>
<dd>30</dd>
<dt>lims</dt>
<dd>the bounded 1D domain on which to set up the bisquare functions</dd>
<dt>dtype</dt>
<dd>data type</dd></dl>

Arguments

Bisquare functions on a 2D domain — bisquares2D

<dl>

<dt>r</dt>
<dd>30</dd>


<dt>lims</dt>
<dd>the bounded 1D domain on which to set up the bisquare functions</dd>


<dt>dtype</dt>
<dd>data type</dd>

</dl>

bisquares2D: Bisquare functions on a 2D domain

Description

Usage

Value

Arguments