-
Notifications
You must be signed in to change notification settings - Fork 0
/
01_classes.R
273 lines (205 loc) · 8.97 KB
/
01_classes.R
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
# Fields class -------
setClass('fieldspec',
# define slots and respective types
slots = list(
# Admin -------------------- #
specsnr = 'integer' , # specs row of /fields/settings.RData
parsetting = 'character' , # high/low
gaussian = 'logical' , # is Gaussian? T/F
extra_par = 'vector' , # non-gaussian field extra-parameters
fielditer_n = 'integer' , # number of iterations for the same field. Preset to 200
fielditer = 'integer' , # iteration number
seed = 'integer' , # seed used to generate this field (master_seed_number * fielditer)
# Field parameters --------- #
nsize = 'integer' , # size of the field. Filled later on based on parsetting
gridded = 'logical' , # if TRUE = gridded , FALSE = ssh
min_dist = 'numeric' , # if gridded = FALSE, min distance between locations
domain = 'vector' , # 2x2 matrix with boundaries for the domain
truetheta = 'vector' , # true 4 length vector thetas (range, psill, smoothness, nugget)
truebetas = 'vector' , # true 3 length vector of betas (beta0, beta_x, beta_y)
# Store info --------------- #
locs = 'matrix' , # expand.grid output
fulldata = 'vector' # complete matern simulation observations
),
prototype = list(
# Admin -------------------- #
specsnr = NA_integer_ ,
parsetting = NA_character_ ,
gaussian = TRUE ,
extra_par = vector() ,
fielditer_n = 200L ,
fielditer = NA_integer_ ,
seed = NA_integer_ ,
# Field parameters --------- #
nsize = NA_integer_ ,
gridded = T ,
min_dist = 0.01 ,
domain = matrix(c(0,1, 0,1), nrow = 2, ncol = 2, byrow = TRUE) ,
truetheta = vector() ,
truebetas = c(1, 0.1, 0.2) ,
# store info --------------- #
locs = matrix() ,
fulldata = vector()
)
)
# Define validation function for the Field class ----
setGeneric(name = 'validate',
def = function(theObject, i_ref) {
standardGeneric('validate') })
setMethod(f = 'validate',
signature = 'fieldspec',
definition = function(theObject, i_ref) {
# update seed and rep
theObject@fielditer <- as.integer(i_ref)
theObject@seed <- as.integer(i_ref * seed_master_number)
set.seed(seed = theObject@seed) # reproducibility
# par_setting
if(theObject@parsetting == 'low'){
theObject@truetheta <- c(0.05, 1, 0.5, 0.1)
} else {
if(theObject@parsetting == 'high'){
theObject@truetheta <- c(0.05, 1, 2.3, 0)
} else stop('parsetting argument not correct')
}
# sampling_design design
if(theObject@gridded){ # Gridded
x_seq <- seq(theObject@domain[1,1], theObject@domain[1,2], length.out = sqrt(theObject@nsize))
y_seq <- seq(theObject@domain[2,1], theObject@domain[2,2], length.out = sqrt(theObject@nsize))
theObject@locs <- as.matrix(expand.grid(x = x_seq, y = y_seq))
} else{ # SSI
aux_domain <- owin(c(theObject@domain[1,1], theObject@domain[1,2]), c(theObject@domain[2,1], theObject@domain[2,2]))
aux_rSSI <- rSSI(r = theObject@min_dist, n = theObject@nsize, win = aux_domain)
locs <- cbind('x_seq' = aux_rSSI[[3]],'y_seq' = aux_rSSI[[4]])
theObject@locs <- locs[order( locs[,1]), ]
}
# simulate field
if (theObject@gaussian){ # then gaussian
# complete distance matrix
distmat <- as.matrix(dist(theObject@locs))
Sigma <- cov.mat2(h = distmat, theta = theObject@truetheta)
rm(distmat)
# gc(verbose = FALSE, reset = T)
iidsample <- rnorm(theObject@nsize)
cholS <- chol(Sigma)
partial <- theObject@truebetas[1] + theObject@truebetas[2] * cos(theObject@locs[,1]) +
theObject@truebetas[3]*cos(theObject@locs[,2])
theObject@fulldata <- partial + as.vector(iidsample %*% cholS)
} else {
theObject@extra_par <- c(0.5, 0.8) # skew, tail (kurtosis)
# Simulation of the spatial SAS RF:
tmp_data <- GeoSim(coordx = theObject@locs, corrmodel = 'Matern', model = 'SinhAsinh',
param = list(skew = theObject@extra_par[1], tail = theObject@extra_par[2],
smooth = theObject@truetheta[3], mean = 0,
sill = theObject@truetheta[2], scale = theObject@truetheta[1],
nugget = theObject@truetheta[4]))$data
# Outputs
partial <- theObject@truebetas[1] + theObject@truebetas[2] * cos(theObject@locs[,1]) +
theObject@truebetas[3]*cos(theObject@locs[,2])
theObject@fulldata <- partial + tmp_data
}
return(theObject)
}
)
# Estimation classes --------
# 03a full matern class -----
setClass('matern_estimate',
# define slots and respective types
slots = list(
# admin
specsnr = 'integer',
fielditer = 'integer',
# subsample
time = 'numeric',
optimout = 'ANY',
betas_hat = 'vector',
memory_used = 'numeric',
loadstring = 'character'),
prototype = list(
# admin
specsnr = NA_integer_,
fielditer = NA_integer_,
# subsample
time = NA_real_,
optimout = NULL,
betas_hat = vector(),
memory_used = NA_real_,
loadstring = NA_character_)
)
# 03b tapering class -----
setClass('taper_estimate',
# define slots and respective types
slots = list(
# admin
specsnr = 'integer',
fielditer = 'integer',
delta = 'numeric',
# subsample
time = 'numeric',
optimout = 'ANY',
betas_hat = 'vector',
memory_used = 'numeric',
loadstring = 'character'),
prototype = list(
# admin
specsnr = NA_integer_,
fielditer = NA_integer_,
delta = NA_real_,
# subsample
time = NA_real_,
optimout = NULL,
betas_hat = vector(),
memory_used = NA_real_,
loadstring = NA_character_)
)
# 03c directmiss class -----
setClass('directmiss_estimate',
# define slots and respective types
slots = list(
# admin
specsnr = 'integer',
fielditer = 'integer',
delta = 'numeric',
# subsample
time = 'numeric',
optimout = 'ANY',
betas_hat = 'vector',
memory_used = 'numeric',
loadstring = 'character'),
prototype = list(
# admin
specsnr = NA_integer_,
fielditer = NA_integer_,
delta = NA_real_,
# subsample
time = NA_real_,
optimout = NULL,
betas_hat = vector(),
memory_used = NA_real_,
loadstring = NA_character_)
)
# 03d composite class -----
setClass('composite_estimate',
# define slots and respective types
slots = list(
# admin
specsnr = 'integer',
fielditer = 'integer',
delta = 'numeric',
# subsample
time = 'numeric',
optimout = 'ANY',
betas_hat = 'vector',
memory_used = 'numeric',
loadstring = 'character'),
prototype = list(
# admin
specsnr = NA_integer_,
fielditer = NA_integer_,
delta = NA_real_,
# subsample
time = NA_real_,
optimout = NULL,
betas_hat = vector(),
memory_used = NA_real_,
loadstring = NA_character_)
)