
Wrapper function to generate PHATE parameters
Source:R/params-examples.R, R/params-generated.R
params_phate.RdWrapper function to generate PHATE parameters
Usage
params_phate(
decay = 40,
bandwidth_scale = NULL,
t_max = 100L,
t_custom = NULL,
gamma = 1,
thresh = 1e-04,
n_landmarks = 2048L,
landmark_method = c("spectral", "random", "density"),
n_svd = NULL,
mds_method = c("sgd_dense", "classic"),
mds_iter = NULL,
graph_symmetry = c("additive", "multiplicative", "mnn", "none")
)Arguments
- decay
Numeric. Alpha decay parameter controlling the kernel bandwidth. Higher values create sharper transitions. Defaults to
40.0.- bandwidth_scale
Numeric or
NULL. Scaling factor applied to the kernel bandwidth. IfNULL, uses1.0. Defaults toNULL.- t_max
Integer. Maximum diffusion time considered during automatic selection via Von Neumann entropy knee point detection. At least
3, as the knee search needs three points. Defaults to100L.- t_custom
Integer or
NULL. Fixed diffusion time. If set, overrides automatic time selection. Defaults toNULL.- gamma
Numeric. Informational distance parameter for the diffusion potential.
1is the log potential. Defaults to1.0.- thresh
Numeric. Sparsity threshold on the kernel. Neighbours past
kjoin the kernel while their affinity clears it. Defaults to1e-04.- n_landmarks
Integer or
NULL. Number of landmarks for compressed diffusion. IfNULL, or at least the number of samples, the full N x N diffusion operator is used. Defaults to2048L.- landmark_method
String. Method used to select landmarks. One of
c("spectral", "random", "density"). Defaults to"spectral".- n_svd
Integer or
NULL. Number of singular vectors for spectral landmark selection. IfNULL, uses100. Ignored by the other landmark methods. Defaults toNULL.- mds_method
String. MDS algorithm used for the final embedding. One of
c("sgd_dense", "classic"). Defaults to"sgd_dense".- mds_iter
Integer or
NULL. Number of SGD-MDS iterations. IfNULL, scales with the number of points:300below 1000,500below 5000 and800otherwise. Defaults toNULL.- graph_symmetry
String. Method used to symmetrise the affinity graph. One of
c("additive", "multiplicative", "mnn", "none"). Defaults to"additive".
Value
A named list with the following elements:
decay - Numeric. Alpha decay parameter controlling the kernel bandwidth. Higher values create sharper transitions. Defaults to
40.0.bandwidth_scale - Numeric or
NULL. Scaling factor applied to the kernel bandwidth. IfNULL, uses1.0. Defaults toNULL.t_max - Integer. Maximum diffusion time considered during automatic selection via Von Neumann entropy knee point detection. At least
3, as the knee search needs three points. Defaults to100L.t_custom - Integer or
NULL. Fixed diffusion time. If set, overrides automatic time selection. Defaults toNULL.gamma - Numeric. Informational distance parameter for the diffusion potential.
1is the log potential. Defaults to1.0.thresh - Numeric. Sparsity threshold on the kernel. Neighbours past
kjoin the kernel while their affinity clears it. Defaults to1e-04.n_landmarks - Integer or
NULL. Number of landmarks for compressed diffusion. IfNULL, or at least the number of samples, the full N x N diffusion operator is used. Defaults to2048L.landmark_method - String. Method used to select landmarks. One of
c("spectral", "random", "density"). Defaults to"spectral".n_svd - Integer or
NULL. Number of singular vectors for spectral landmark selection. IfNULL, uses100. Ignored by the other landmark methods. Defaults toNULL.mds_method - String. MDS algorithm used for the final embedding. One of
c("sgd_dense", "classic"). Defaults to"sgd_dense".mds_iter - Integer or
NULL. Number of SGD-MDS iterations. IfNULL, scales with the number of points:300below 1000,500below 5000 and800otherwise. Defaults toNULL.graph_symmetry - String. Method used to symmetrise the affinity graph. One of
c("additive", "multiplicative", "mnn", "none"). Defaults to"additive".
Examples
params_phate()
#> $decay
#> [1] 40
#>
#> $bandwidth_scale
#> NULL
#>
#> $t_max
#> [1] 100
#>
#> $t_custom
#> NULL
#>
#> $gamma
#> [1] 1
#>
#> $thresh
#> [1] 1e-04
#>
#> $n_landmarks
#> [1] 2048
#>
#> $landmark_method
#> [1] "spectral"
#>
#> $n_svd
#> NULL
#>
#> $mds_method
#> [1] "sgd_dense"
#>
#> $mds_iter
#> NULL
#>
#> $graph_symmetry
#> [1] "additive"
#>
params_phate(decay = 20, n_landmarks = 500L)
#> $decay
#> [1] 20
#>
#> $bandwidth_scale
#> NULL
#>
#> $t_max
#> [1] 100
#>
#> $t_custom
#> NULL
#>
#> $gamma
#> [1] 1
#>
#> $thresh
#> [1] 1e-04
#>
#> $n_landmarks
#> [1] 500
#>
#> $landmark_method
#> [1] "spectral"
#>
#> $n_svd
#> NULL
#>
#> $mds_method
#> [1] "sgd_dense"
#>
#> $mds_iter
#> NULL
#>
#> $graph_symmetry
#> [1] "additive"
#>