sfdep 0.2.3
CRAN release: 2023-01-11
- remove readr suggestion to pass
noSuggestsfrom CRAN. Much minuscule tasks for the overlords.
sfdep 0.2.1
CRAN release: 2022-12-16
- Bivariate Moran uses the new spdep implementation which is faster. It does, however, return an object of class
bootwhich needs extra work to calculate the p-value - Local join count univariate uses spdep now
- adds graph based neighbors:
- st_nb_delaunay() uses spdep::tri2nb()
- st_nb_gabriel() uses spdep::gabrielneigh() and spdep::graph2nb()
- st_nb_relative() uses spdep::relativeneigh() and spdep::graph2nb()
-
activate()is now exported as a generic method to no longer conflict with tidygraph -
st_as_graph()now creates undirected graphs - new functions for sfnetworks
-
node_get_nbs(): converts the node adjacency list to a neighbor list -
node_get_edge_list(): creates an edge adjacency list used to get edge attribute columns -
node_get_edge_col(): uses the edge list to get edge attributes and turn them into a weight like list
-
sfdep 0.2.0
CRAN release: 2022-08-15
-
spatial_gini()is a new global measure -
euclidean_median()is added for identifying the euclidean median of points for point pattern analysis.- Implemented using
pracma::geo_median()-
pracmais now added as a suggested package
-
- Implemented using
-
center_mean()andcenter_median()are added for point pattern analysis -
tally_jc()returns join count matrix viaspdep::joincount.multi()as a data.frame object -
st_complete_nb()creates a complete graph of neighbors - set operations for neighbor lists
nb_union(),nb_intersect(),nb_setdiff() -
st_block_nb()is introduced to create neighbor contiguity based on spatial regimes -
critical_threshold()gains argumentkto identify minimum number of neighbors when calculating threshold -
pct_nonzero()calculates the percent of non-zero neighbors -
szero()to calculate the global sum of weights -
nb_as_matrix()andwt_as_matrix()convert neighbor and weights lists to matrices - two new functions
include_self()based onspdep::include.self()and newremove_self()which is available separately in spdep due to issue 83 asspdep::remove.self(). -
local_g_perm()returns correct G statistic. Previously returned only the internal measure. -
local_gstar()andlocal_gstar_perm()do not warn whenself.included = TRUE -
emerging_hotspot_analysis()implements emerging hotspot analysis with spacetime objects - a new
spacetimeclass for linking spatio-temporal data and geometry- see the spacetime vignette.
- Implemented colocation quotient (CLQ) measures See the colocation vignette:
-
global_colocation()for global colocation of one categorical variable -
pairwise_colocation()for asymmetric colocation of subcategories of one or two categories -
local_colocation()for local CLQ measure - these functions are not fast! PRs welcomed to improve computation speed.
-
-
st_kernel_weights()now returns an attributekernelwhich is set to the value of the kernel argument -
cond_permute_nb()specifiesSIMPLIFY = FALSEin internalmapply()call which was causing errors in creating conditionally permuted neighbor lists