pkgsrc-Changes archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
CVS commit: pkgsrc/geography/R-spatstat.explore
Module Name: pkgsrc
Committed By: mef
Date: Fri Jun 12 13:54:30 UTC 2026
Modified Files:
pkgsrc/geography/R-spatstat.explore: Makefile distinfo
Log Message:
(geography/R-spatstat.explore) Update 3.3.4 to 3.8.1
CHANGES IN spatstat.explore VERSION 3.8-1
OVERVIEW
o We thank Mario D'Antuono and Hamid Ghorbani for contributions.
o Reverted to previous default choice of estimator in pair correlation.
o Changed rule for converting bandwidths in pair correlation.
o Improved handling of boundary pixels in density(), relrisk(), Smooth()
o Improvements to rose diagrams.
o Fiksel's bandwidth rule.
o Faster calculation of Ripley's isotropic edge correction.
o Minor improvements and bug fixes.
NEW FUNCTIONS
o bw.fiksel
Fiksel's (1988) rule for smoothing bandwidth for
estimating the pair correlation function.
SIGNIFICANT USER-VISIBLE CHANGES
o pair correlation functions
(pcf.ppp, pcfinhom, pcfcross, pcfdot, pcfmulti,
pcfcross.inhom, pcfdot.inhom, pcfmulti.inhom)
The default value of 'divisor' has been changed to "r"
(reverting to the behaviour in spatstat.explore 3.7-0 and earlier)
until some technical problems are resolved.
o pair correlation functions
(pcf.ppp, pcfinhom, pcfcross, pcfdot, pcfmulti,
pcfcross.inhom, pcfdot.inhom, pcfmulti.inhom)
The rule for converting bandwidth 'bw' from the distance scale to the
area scale (when divisor="a") has changed.
o pcfinhom
Argument 'adjust' has been renamed 'adjust.bw' to avoid confusion.
However 'adjust' is still accepted.
o density.ppp, density.psp, relrisk.ppp, Smooth.ppp
These methods now accept the argument 'rule.pix' passed to 'owin2mask'.
This makes it possible to ensure that boundary pixels which partially
intersect the window are included in the resulting pixel image,
by setting op="cover".
o rose.default, rose.histogram, rose.density, rose.fv, circdensity
Rose diagrams can now be computed and plotted for data
expressed in any units with any period (such as hours in a 24-hour day).
More options for argument 'unit'.
New argument 'fullcircle'.
o rose.default, rose.histogram
Polygons are now filled, by default, in light grey.
New formal argument 'col' specifies the fill colour(s).
o rose.default, rose.histogram, rose.density, rose.fv
New argument 'do.ticks' controls whether tick marks are drawn.
o rose.default, rose.histogram, rose.density, rose.fv
Concentric rings are now drawn inside the circle, for scale.
New argument 'do.rings'.
o rose.default, rose.histogram, rose.density, rose.fv
The return value now has attribute "R" giving the
radius of the outer circle of the diagram, and attribute "rings"
giving the radii of the concentric scale rings.
o rose.fv
Now plots multiple curves, if there are multiple columns of values.
New argument 'fmla' controls which columns of values are plotted.
o rose.fv
Now handles envelopes (objects of class 'envelope') as well.
o rose.default, rose.histogram, rose.density, rose.fv
New argument 'root' specifies whether to apply a square root transformation.
o Summary functions
Calculation of Ripley's isotropic edge correction has been accelerated,
and responds more quickly to user interrupts.
o edge.Ripley
New optional argument 'bdistX'.
o edge.Ripley
Responds more quickly to user interrupts
(STOP command, Ctrl-C signal).
BUG FIXES
o pcfdot, pcfcross, pcfdot.inhom, pcfcross.inhom
Envelopes for these functions were sometimes incorrect in the sense that
the summary functions were not all calculated with the same settings.
[Spotted by Hamid Ghorbani.]
Affected spatstat.explore 3.8-0 only.
Fixed.
o plot.fv
Occasionally gave a strange warning about 'distmap' methods.
Fixed.
o rose.default, rose.histogram
Crashed if argument 'labels' was given.
[Spotted by Mario D'Antuono.]
Fixed.
o rose.default, rose.histogram
Geometry of plot was sometimes incorrect when clockwise=FALSE.
[Spotted by Mario D'Antuono.]
Fixed.
o rose.default, rose.histogram
Ignored arguments 'freq' and 'probability'.
Fixed.
o rose.default, rose.histogram, rose.density, rose.fv
Argument 'do.plot=FALSE' was partially ignored (i.e. a graphics device
was opened and some graphical commands were issued).
Fixed.
CHANGES IN spatstat.explore VERSION 3.8-0
OVERVIEW
o We thank Andrew Bevan, Meri Bilan, Tilman Davies,
Andrea Gilardi and Martin Hazelton for contributions.
o Important changes to default settings for pair correlation functions.
o More flexible interpretation of arguments 'lambda = <ppm>"
in inhomogeneous summary statistics
o New options for multitype pair correlation functions.
o More options for 'simulate' argument in 'envelope' methods.
o Specify the 'theoretical' value of the summary function in 'envelope'
o Extended capability of 'envelope.pp3'
o Method of T.F. Cox for identifying dense and sparse regions of points.
o Minor improvements and bug fixes.
NEW FUNCTIONS
o coxmap
Given a point pattern, divide the spatial region into
areas where the pattern is dense, sparse or neither,
using the method of T.F. Cox (1979).
SIGNIFICANT USER-VISIBLE CHANGES
o pcf.ppp, pcfinhom
The default value of 'divisor' has changed to "a",
so that the new area-based estimator is the default.
o pcf.ppp, pcfinhom
The default value of 'zerocor' is now 'JonesFoster' for small patterns
and 'convolution' for larger patterns.
o pcf.ppp, pcfinhom
New argument 'nsmall' determines the threshold for large patterns.
o pcfinhom
Default value of 'normpower' has changed to 2.
o pcfcross, pcfdot, pcfmulti
The default estimation method has changed. The new default is
based on squared distances (divisor="a").
o pcfcross.inhom, pcfdot.inhom, pcfmulti.inhom
The default estimation method has changed. The new default is
based on squared distances (divisor="a").
o pcfmulti, pcfmulti.inhom
New arguments h, bw.args, adjust, zerocor, nsmall, gref, tau, close.
o pcfcross.inhom, pcfdot.inhom, pcfmulti.inhom,
Interpretation of arguments 'lambdaI', 'lambdaJ', 'lambdaX', 'lambdadot'
has changed when they are fitted point process models. If the model is
a multitype point process then it is assumed to be a model for the
entire point pattern X, while if the model is an unmarked point process
then it is assumed to be a model for the relevant subset of the data.
o Kcross.inhom, Kdot.inhom, Kmulti.inhom
Interpretation of arguments 'lambdaI', 'lambdaJ', 'lambdaX', 'lambdadot'
has changed when they are fitted point process models. If the model is
a multitype point process then it is assumed to be a model for the
entire point pattern X, while if the model is an unmarked point process
then it is assumed to be a model for the relevant subset of the data.
o pcfcross, pcfdot
Shortened list of formal arguments (other arguments pass to 'pcfmulti')
o pcfcross.inhom, pcfdot.inhom
Shortened list of formal arguments
(other arguments are passed to 'pcfmulti.inhom')
o density.ppp
Argument 'weights' can be logical valued.
o intensity.ppp
Argument 'weights' can be logical valued.
o markcorr, markcrosscorr
Argument 'weights' can be logical valued.
o relrisk.ppp
Argument 'weights' can be logical valued.
o envelope.ppp
Argument 'simulate' can be a point process model of class
'ppm', 'kppm', 'dppm', 'slrm', 'clusterprocess' or 'detpointprocfamily'.
o envelope.pp3
Marked point patterns are now accepted.
o envelope.pp3
New arguments 'fix.n' and 'fix.marks' determine whether the
simulated patterns have a fixed or random number of points.
o envelope.ppp, envelope.pp3
New argument 'theoryfun' can be used to provide the theoretical value
of the summary function.
o hopskel, hopskel.test
Accelerated.
BUG FIXES
o quadrat.test
Crashed when argument 'lambda' was an image or function on a linear network.
[Spotted by Andrea Gilardi.]
Fixed.
CHANGES IN spatstat.explore VERSION 3.7-0
OVERVIEW
o Important changes to mark correlation function.
o Summary functions have argument 'rmax'.
o Minor internal improvements.
SIGNIFICANT USER-VISIBLE CHANGES
o markcorr
The interpretation of the argument 'weights' has changed.
o markcorr
New argument 'fadjust' (equivalent to the old 'weights').
o markcrosscorr
New argument 'weights'.
o plot.fasp
Now returns a list containing the results of plot.fv for each panel.
o summary functions
The following functions have a new formal argument 'rmax':
Fest, Finhom, FmultiInhom,
Gcross, Gdot, Gest, Ginhom, Gmulti, GmultiInhom,
Jest, Jcross, Jdot, Jinhom, Jmulti,
Kcross, Kdot, Kinhom, Kmulti,
Kcross.inhom, Kdot.inhom, Kmulti.inhom, Kmark, Ksector,
markconnect, markcorr, markcorrint, markcrosscorr, markequal, markvario,
pcf.ppp, pcfcross, pcfdot, pcfmulti,
pcfinhom, pcfcross.inhom, pcfdot.inhom, pcfmulti.inhom.
BUG FIXES
o plot.fv
Crashed sometimes if the formula involved the name of
an object that was not a column of x.
Fixed.
CHANGES IN spatstat.explore VERSION 3.6-0
OVERVIEW
o We thank Lucia Cobo Sanchez and Tilman Davies for contributions.
o Nearest neighbour counting functions.
o Tolerance contours for relative risk.
o Minor improvements and bug fixes.
NEW FUNCTIONS
o markequal
Mark equality function e(r) for multitype point patterns.
o nnequal, nncount
Summary functions for multitype point patterns
based on the types of nearest neighbours.
o tolcon
Tolerance contours for spatial relative risk.
SIGNIFICANT USER-VISIBLE CHANGES
o markcorr, markconnect, markvario, Emark, Vmark
The result now has an attribute 'smooth.args'
containing the smoothing parameters used.
BUG FIXES
o Kmark
Calculation was incorrect for multitype patterns (categorical marks).
Fixed.
o as.data.frame.fv
Some attributes of x were incorrectly retained.
Fixed.
CHANGES IN spatstat.explore VERSION 3.5-3
OVERVIEW
o Summary functions handle NA objects.
SIGNIFICANT USER-VISIBLE CHANGES
o Summary functions
(Kest, Kest.fft, Kcross, Kdot, Kmulti, Kscaled,
Kinhom, Kcross.inhom, Kdot.inhom, Kmulti.inhom,
Lest, Lcross, Ldot, Linhom, Lscaled, Lcross.inhom, Ldot.inhom,
pcf, pcfcross, pcfdot, pcfmulti, pcfinhom, pcfcross.inhom, pcfdot.inhom,
Fest, Gest, Gcross, Gdot, Gmulti, Jest, Jcross, Jdot, Jmulti, Iest,
Finhom, Ginhom, Jinhom, markcorr, markvario, markconnect, Emark, Vmark)
If X is a missing point pattern, a missing object of class 'fv' is returned.
o Local summary functions
(localL, localK, localLcross, localKcross, localLdot, localKdot
localLcross.inhom, localKcross.inhom, localLdot.inhom, localKdot.inhom)
If X is a missing point pattern, a missing 'data.frame' is returned.
o allstats, alltypes
If the first argument is a missing point pattern,
a missing object of class 'fasp' is returned.
o Kmeasure
If X is a missing point pattern, a missing pixel image is returned.
CHANGES IN spatstat.explore VERSION 3.5-2
OVERVIEW
o Internal fixes.
CHANGES IN spatstat.explore VERSION 3.5-1
OVERVIEW
o Tweak documents to satisfy CRAN.
CHANGES IN spatstat.explore VERSION 3.5-0
OVERVIEW
o We thank Jakob Jentschke and Suman Rakshit for contributions.
o Extension of ROC curves.
o Shrinkage estimate in Smooth.ppp.
o Youden index.
o Minor improvements and bug fixes.
NEW FUNCTIONS
o roc.rhohat, roc.cdftest, roc.bermantest
Calculate ROC from other kinds of data.
o roc.im
Calculate ROC from a pixel image such as an estimate of intensity.
o youden
Youden statistic for ROC curves.
SIGNIFICANT USER-VISIBLE CHANGES
o cov.im, cor.im
Default changed to 'use="complete.obs"'.
o roc.ppp
New argument 'baseline' allows calculation of ROC relative to a baseline.
o roc.ppp
New argument 'weights' for numerical weights on the data points.
o roc.ppp
New argument 'observations' specifies whether to use the exact point
coordinates or pixel presence-absence indicators.
o roc.ppp
New argument 'method' determines the estimation method.
New options include a kernel smoothing estimate and a monotone estimate.
o roc.ppp
New arguments 'CI', 'alpha' for calculating confidence intervals.
o roc.ppp
New argument 'subset' specifies a sub-region of the spatial domain
in which the ROC should be calculated.
o Smooth.ppp
Shrinkage estimate implemented.
New arguments 'shrink' and 'shrinktype'.
o Smooth.ppp
Slight change to algorithm for handling very small bandwidths.
o SpatialQuantile.ppp, SpatialMedian.ppp
Slight change to algorithm for handling very small bandwidths.
CHANGES IN spatstat.explore VERSION 3.4-3
OVERVIEW
o Shrinkage estimator of relative risk.
o Normalised relative risk.
o Extensions to cross-validated bandwidth selection for smoothing.
o Bug fixes.
SIGNIFICANT USER-VISIBLE CHANGES
o relrisk.ppp
New argument 'shrink' supports Bithell's shrinkage estimator.
o relrisk.ppp
New argument 'normalise' controls whether the estimate will be
normalised so that constant risk corresponds to the value 1.
o bw.smoothppp
New arguments 'train' and 'test' for cross-validation.
BUG FIXES
o Smooth.ppp
The estimate of standard error could sometimes take the value NaN.
Fixed.
o Smooth.ppp
Terminated in some cases with error message "oo not found".
Fixed.
CHANGES IN spatstat.explore VERSION 3.4-2
OVERVIEW
o Minor fix.
CHANGES IN spatstat.explore VERSION 3.4-1
OVERVIEW
o Minor change to documentation to satisfy CRAN package checker.
CHANGES IN spatstat.explore VERSION 3.4-0
OVERVIEW
o We thank Ya-Mei Chang, Tilman Davies, Martin Hazelton
and Tingting Zhan for contributions.
o New estimator of pair correlation function.
o New methods for bandwidth selection for pair correlation estimation.
o Helper functions for measuring estimator performance.
NEW FUNCTIONS
o bw.bdh
Adjust bandwidth of kernel estimate of pair correlation function,
to account for inhomogeneity.
o bw.pcfinhom
Cross-validation rule for selecting bandwidth of kernel estimate
of (inhomogeneous) pair correlation function.
o ptwise.envelope, bias.envelope, RMSE.envelope
Calculate pointwise statistics of the simulated function values
in an envelope object.
o ISE.envelope, ISB.envelope, IV.envelope
Integrated squared error, integrated squared bias, integrated variance
of the simulated function values in an envelope object.
o MISE.envelope
Mean integrated squared error of the simulated function values
in an envelope object.
SIGNIFICANT USER-VISIBLE CHANGES
o pcf.ppp
New option: divisor="a" calculates a new estimator of the pcf
with improved performance.
o pcfinhom
New option: divisor="a" calculates a new estimator of the inhomogeneous pcf
with improved performance.
o bw.stoyan
New argument 'extrapolate' supports a modification of Stoyan's rule.
o plot.fv
New argument 'do.plot'.
To generate a diff of this commit:
cvs rdiff -u -r1.3 -r1.4 pkgsrc/geography/R-spatstat.explore/Makefile \
pkgsrc/geography/R-spatstat.explore/distinfo
Please note that diffs are not public domain; they are subject to the
copyright notices on the relevant files.
Modified files:
Index: pkgsrc/geography/R-spatstat.explore/Makefile
diff -u pkgsrc/geography/R-spatstat.explore/Makefile:1.3 pkgsrc/geography/R-spatstat.explore/Makefile:1.4
--- pkgsrc/geography/R-spatstat.explore/Makefile:1.3 Sat Feb 8 09:32:21 2025
+++ pkgsrc/geography/R-spatstat.explore/Makefile Fri Jun 12 13:54:30 2026
@@ -1,7 +1,7 @@
-# $NetBSD: Makefile,v 1.3 2025/02/08 09:32:21 mef Exp $
+# $NetBSD: Makefile,v 1.4 2026/06/12 13:54:30 mef Exp $
R_PKGNAME= spatstat.explore
-R_PKGVER= 3.3-4
+R_PKGVER= 3.8-1
CATEGORIES= geography
MAINTAINER= pkgsrc-users%NetBSD.org@localhost
Index: pkgsrc/geography/R-spatstat.explore/distinfo
diff -u pkgsrc/geography/R-spatstat.explore/distinfo:1.3 pkgsrc/geography/R-spatstat.explore/distinfo:1.4
--- pkgsrc/geography/R-spatstat.explore/distinfo:1.3 Sat Feb 8 09:32:21 2025
+++ pkgsrc/geography/R-spatstat.explore/distinfo Fri Jun 12 13:54:30 2026
@@ -1,5 +1,5 @@
-$NetBSD: distinfo,v 1.3 2025/02/08 09:32:21 mef Exp $
+$NetBSD: distinfo,v 1.4 2026/06/12 13:54:30 mef Exp $
-BLAKE2s (R/spatstat.explore_3.3-4.tar.gz) = 3bff7d879cfcf146f5a335dacb6008c7aa3af940494fd1b88b496d487bec4421
-SHA512 (R/spatstat.explore_3.3-4.tar.gz) = 8e2ad136f42a96ad52b310a96b26bf640c4f219d54aadba5f65bc29f40a7bfee073fa919313cbe0b643e30978b8f8100d52969f817ab73c845db68b8f3cd1e84
-Size (R/spatstat.explore_3.3-4.tar.gz) = 668417 bytes
+BLAKE2s (R/spatstat.explore_3.8-1.tar.gz) = ca69a077940415a865876a3e280242631c20fd7d26e546cc7f0626f692a0a9bb
+SHA512 (R/spatstat.explore_3.8-1.tar.gz) = 02841fccd54df1c9320c1cf4cb4171a229ed1a4f524509057e7319f889862ac23687f428eb674cb6338273955f97ebc8832fa966ce33c99c06c7de539950a044
+Size (R/spatstat.explore_3.8-1.tar.gz) = 739496 bytes
Home |
Main Index |
Thread Index |
Old Index