-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathNEWS
348 lines (255 loc) · 12.8 KB
/
NEWS
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
Changes in version 2.9.3 (2024-09-28)
New features or changed behaviour of old functions
o none
Changes in version 2.9.2 (2024-09-27)
New features or changed behaviour of old functions
o minsimRSE new method for optimalSpacing simulations
Bugs fixed
o run.scenarios() with fit="multifit" failed with default (shared) mask
o estimateSummary() lost scenario number in some grouped examples
o run.scenarios() with model2D = 'IHP' could fail with NA start value for D
Other
o internal function dfcast accepts character detectfn
o plot.optimalSpacing tweaked to allow pch other than 1 and new plottype option
o run.scenarios() with trapset functions outputs a message as each is tested
o added anchors to external documentation links
Changes in version 2.9.1 (2024-04-20)
New features or changed behaviour of old functions
o run.scenarios() default extractfn keeps raw counts (including RPSV) as attribute
o run.scenarios() acquires argument 'prefix' for saving intermediate output
o expand.arg() new utility function to generate lists of arguments
o scenarioSummary() outputs overlap index k
Bugs fixed
o estimateSummary() failed when some estimates NA
Other
o run.scenarios() tweak to allow missing detectpar in det.pars
o estimateArray() and estimateSummary() more robust to failed model fit
Changes in version 2.9.0 (2024-03-01)
New features or changed behaviour of old functions
o multi-model fits (nested fit.args for one fitindex)
o run.scenarios() acquires argument 'joinsessions'
Bugs fixed
o run.scenarios() could ignore maskset
o run.scenarios() failed with byscenario = TRUE
o Enrm() and related functions: E(m) sometimes NA due to near-zero hazard
Other
o outputtype<- in-place replacement (also reassigns class)
o transformOutput() function to recast output
o select.stats tweak to allow movingAverage
o estimateArray() and estimateSummary() more robust
o byscenario = TRUE uses cluster type = FORK on Unix-like machine
o tweaks to c.estimatetables and c.selectedstatistics()
o run.scenarios() fitarg start may be NULL (dodges binomial problem with "true")
o onAttach() options (matprod = "internal")
Changes in version 2.8.2 (2023-03-11)
New features or changed behaviour of old functions
o none
Other
o example time reduced for CRAN
Changes in version 2.8.1 (2023-03-10)
New features or changed behaviour of old functions
o fit option 'design' (assumes detection model cf Fewster & Buckland 2004)
o estimateArray() and estimateSummary() new functions
Other
o summary field 'var'
Changes in version 2.8.0 (2023-01-19)
New features or changed behaviour of old functions
o run.scenarios() fit.function may be "ipsecr.fit"
o run.scenarios() acquires argument 'CH.function'
o functions to combine output from different executions
of run.scenarios() (methods c, rbind for estimatetables, selectedstatistics)
Other
o suggests ipsecr
o schematic figure in help page ?secrdesign
o revised help page ?run.scenarios
o run.scenarios() sets RCPP_PARALLEL_NUM_THREADS if ncores provided
Changes in version 2.7.0 (2022-12-04)
New features or changed behaviour of old functions
o GAoptim() optimization by genetic algorithm using min(n,r) criterion
(Durbach et al. 2021) and others
o minnrRSE() accepts GAoptim() output
o En2() and Qpm() for additional optimization criteria
o run.scenarios() option to create trapset dynamically with function(s)
o run.scenarios() compatible with dynamic density function in pop.args
Bugs fixed
o run.scenarios() failed with grouped scenarios if maskindex not specified
Other
o on GitHub
o added tests
o exports internal functions GApenfn() and compactSample()
o make.scenarios allows single-level 'groups'
o check for consistent traps etc. across groups within scenario
o parallel clusters not used by run.scenarios() if fit = FALSE
o shift to Rcpp, RcppArmadillo from C
o summaryScenarios() ncores defaults to 1 (usually faster than >1)
Changes in version 2.6.0 (2022-01-26)
New features or changed behaviour of old functions
o run.scenarios() 'fit' option uses multithreading capability of secr
o run.scenarios() 'ncores' defaults to NULL (then uses value of environment
variable RCPP_PARALLEL_NUM_THREADS set by setNumThreads())
o run.scenarios() 'byscenario' defaults to FALSE
Bugs fixed
o Enrm() expected movement could fail when some of mask far from detectors
Other
o removed openCR options and imports (deprecated from 2.5.8)
o initial value of RCPP_PARALLEL_NUM_THREADS (usually 2) is set by secr onload
o depends R (>= 3.5.0), secr (>= 4.2.0)
Changes in version 2.5.11 (2020-04-10)
New features or changed behaviour of old functions
o none
Bugs fixed
o run.scenarios() default extractfn (no model fitted) mistake in RSEp approximation
Changes in version 2.5.10 (2020-01-31)
New features or changed behaviour of old functions
o run.scenarios() default extractfn (no model fitted) uses RSEp approximation of
Efford & Boulanger 2019
Bugs fixed
o optimalSpacing() with simulations, fit.function = "openCR.fit" and nrepeats>1
miscalculated RB(D-hat)
o summary.selectedstatistics() failed when run.scenarios() pop.args included poly
o header() stumbled on non-numeric (covariate) specification of D
o run.scenarios() spurious warning for non-numeric D
Bug not fixed (see Other)
o With fit.function 'openCR.fit', the calculation of RB(D-hat) in
summarized simulations does not reflect nrepeats>1
Other
o fit.function 'openCR.fit' is deprecated, and will be removed in a later version;
secr.fit is now fast enough
o optimalSpacing() catches bad root finding when limits inappropriate
o optimalSpacing() only passes valid fit.args to run.scenarios
o force ncores = 1 when fitting model (internal function processCH)
Changes in version 2.5.7 (2019-02-04)
New features or changed behaviour of old functions
o none
Bugs fixed
o count.summary() failed with fit.function = "openCR.fit"
Changes in version 2.5.6 (2019-01-26)
New features or changed behaviour of old functions
o make.scenarios() allows uniform (detectfn = 4)
o count and predict methods for simulation objects inheriting from 'summary'
o select.stats() respects count statistics
Bugs fixed
o plot.optimalSpacing() failed if any values missing (NA)
o optimalSpacing() did not apply nrepeats when computing binomial rotRSE
o scenarioSummary() failed with single-detector arrays
Other
o vignette removed (had been re-introduced by mistake in 2.5.5)
(see www.otago.ac.nz/density/pdfs)
Changes in version 2.5.5 (2018-12-07)
New features or changed behaviour of old functions
o optimalSpacing() new argument 'fit.function' to select openCR.fit/secr.fit
o optimalSpacing() new argument 'distribution'
o optimalSpacing() adjusts rotRSE for binomial distribution of n
o plot.optimalSpacing() does not plot optimum
o summarySpacing() calculates detectors per home range
Bugs fixed
o run.scenarios() tripped up on multi-session simulation
o summaryScenarios() mostly ignored nrepeats
Other
o imports from openCR
Changes in version 2.5.4 (2017-12-05)
New features or changed behaviour of old functions
o none
Other
o vignette secrdesign-vignette.pdf removed (see www.otago.ac.nz/density/pdfs)
Changes in version 2.5.3 (2017-12-04)
New features or changed behaviour of old functions
o saturation() new function for expected detector saturation (trap success)
o scenarioSummary() reports average expected detector saturation
o derived.SL() and regionN.SL() replaced by methods for generic functions
derived and region.N (see 'secr')
Bugs fixed
o Enrm() etc. detectfn 'HAN' and 'HCG' failed
Other
o tweak to select.stats to allow parameters without true value in scenarios
o depends on secr >= 3.1.2
Changes in version 2.5.2 (2017-10-10)
Other
o vignettes secrdesign-Enrm.pdf and secrdesign-tools.pdf removed (see www.otago.ac.nz/density/pdfs)
Changes in version 2.5.0 (2017-10-06)
New features or changed behaviour of old functions
o functions Lambda(), Enrm(), minnrRSE(), costing(), getdetectpar() and optimalSpacing()
o new function scenarioSummary() reports deterministic summary of scenarios (e.g., expected number
of individuals, costings)
o run.scenarios() and fit.models() acquire argument byscenario (parallel scenarios or replicates?)
o fit.models() mask may be passed as component of fit.args, when it has priority over maskset
o run.scenarios() user may specify different masks as core in pop.args
Bugs fixed
o run.scenarios() ignored Nmark detection argument
o chatnsim lost if multiple scenarios
o nrepeats was ignored when method = 'none'
Other
o more robust outputtype when no results
o packageType message if pre-release
o persistent problems in header() for non-scalar arguments
Changes in version 2.4.1 (2016-09-28)
New features or changed behaviour of old functions
o rule-of-thumb RSE added to default extractfn when fit = F
o validate() acquires argument 'quietly'
o detectpar list may be specified in det.args (overrides scenario)
Bugs fixed
o summary.selectedstatistics() failed when type = 'array'
Other
o all array dimensions now named in output of summary.selectedstatistics()
Changes in version 2.4.0 (2016-04-23)
New features or changed behaviour of old functions
o mark-resight data are simulated with secr::sim.resight if traps object
has markocc attribute
o run.scenarios() new argument chatnsim for simulating overdispersion of sightings
o header() allows for complex 'details' and 'start' components
o validate reports number of problem rows
Bugs fixed
o header() failed with model2D="IHP" (problem in argdf - temporary fix)
o run.scenarios() respects detectpar component of det.args (see makeCH)
Other
o number of occasions in markocc overrides noccasions in scenario
o internally streamlined use of multiple cpu (no unnecessary export)
o select.stats() failed with parameter not 'D'
o run.scenarios() uses message instead of cat
Changes in version 2.3.1 (2015-05-14)
New features or changed behaviour of old functions
o scenariosFromStatistics() new function to generate scenarios with specified
number of first captures and recaptures.
o improved output header for fit.args etc.
Changes in version 2.3.0 (2015-03-25)
New features or changed behaviour of old functions
o fit.models() acquires arguments 'scen' and 'repl' to select a subset of rawdata
o run.scenarios() and fit.models() when predict.secr returns a list the default
extractfn selects first session or group, with a warning (previously crashed)
o summary.selectedstatistics() has improved labelling of output
o select.stats() acquires argument 'true' for user override of 'true' values
Bugs fixed
o fit.models() did not work with grouped populations
o default extractfn failed when no detections and fit = FALSE
o run.scenarios() dots argument (...) not passed to worker processes
Other
o more robust handling of start values for secr.fit()
Changes in version 2.2.2 (2015-01-11)
Other
o CITATION file uses package metadata
Changes in version 2.2.1 (2014-11-27)
New features or changed behaviour of old functions
o run.scenarios() allows value "linear" for model2D argument of sim.popn
o details 'userdist' argument suppressed when printing summaries
o make.scenarios() optionally generates scenarios with group structure (e.g. males and females)
o run.scenarios() simulates composite (grouped) populations
Bugs fixed
o validate() did not catch NaN values
Other
o attribute 'regionarea' renamed 'regionsize' to cover linear case
o workflow figure dropped from secrdesign-package.Rd (see vignette)
o argument order changed in summary.selectedstatistics
o default extractfn checks for non-null 'fit' of returned secr object
Changes in version 2.1.3 (2014-06-04)
o vignette renamed and edited
o vignette source code included
Changes in version 2.1.2 (2014-04-29)
New features or changed behaviour of old functions
o validate() function to set out-of-range values to NA
o min, max as fields in summary.selectedstatistics
o secrdesign.pdf vignette
o header in summary output now includes pop.args
o run.scenarios() allows IHP population simulation as
alternative to specifying uniform density
Bugs fixed
o maskset was not saved when specified in run.scenarios()