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
|
#------------------------------------------------------------------------------
# I envision ERD data creation this way
# * STEM_erd.data.creation - processes data from CSV files
# * STEM_erd.data.subsetting - load and further subset data
#
# * Consider SRD & ERD data together as a single functional step
# because they rely on the same parameters
# ==> subsetting should also be done together
#
# Kevin, we will need to talk about how to incorporate
# your work on ERD & NDVI data creation.
# In general we want this to extend to other SRD with
# temporally varying covariates.
#
# Kevin, I dont know where this fits
#------------------------------------------------------------------------------
# Load NDVI Data
#-----------------------------------------------------------------------------
## KFW
## vegetationIndexData <- read.csv(file="ebird_reference_data/processed_VI_Data_SRD.csv", na.strings = c("NA", "?"))
#vegetationIndexData <- matrix(scan("ebird_reference_data/custom_vi.txt"), 18880842, 6, byrow=TRUE)
#print(class(vegetationIndexData))
#print(nrow(vegetationIndexData))
#print(ncol(vegetationIndexData))
testRpyCall <- function(object) {
print(class(object))
print(object)
}
#------------------------------------------------------------------------------
erd.srd.data.subsetting <- function(
erd.data.filename, #erd.data.tag,
srd.pred.design.filename,
spatial.extent.list,
temporal.extent.list,
## KFW speciesList,
species.name,
predictor.names,
response.family,
srd.max.sample.size,
split.par.list,
response.truncation.cutoff=20){
#---------------------------------------------------------------------------
# Load SRD Data #------------------------------------------------------------------------------------------------------------
print(srd.pred.design.filename)
load(file=srd.pred.design.filename) # pred.data
#dim(pred.data$D.pred)
#dim(pred.data$locs)
# ----------------------------------------------------------------------
# Limit Spatial Extent by Point in Rectangle, Polygon, or ShapeFile
# ----------------------------------------------------------------------
xxx <- pred.data$locs$x
yyy <- pred.data$locs$y
region.index <- rep(TRUE, nrow(pred.data$D.pred) )
if (!is.null(spatial.extent.list)){
## KFW if (spatial.extent.list$type == "rectangle"){
if (spatial.extent.list[['type']] == "rectangle"){
ttt.index <- (
yyy > spatial.extent.list$lat.min &
yyy < spatial.extent.list$lat.max &
xxx > spatial.extent.list$lon.min &
xxx < spatial.extent.list$lon.max )
region.index <- (ttt.index & region.index)
}
## KFW if (spatial.extent.list$type == "polygon"){
if (spatial.extent.list[['type']] == "polygon"){
ttt.index <- point.in.polygon(
xxx = xxx,
yyy = yyy,
polygon.vertices =
spatial.extent.list$polygon.vertices)
region.index <- (ttt.index & region.index)
}
## KFW if (spatial.extent.list$type == "shapefile"){
if (spatial.extent.list[['type']] == "shapefile"){
ttt.index <- point.in.shapefile(
sites = data.frame(lon=xxx, lat=yyy),
shape.dir=spatial.extent.list$shape.dir,
shape.filename=spatial.extent.list$shape.filename,
att.selection.column.name=
spatial.extent.list$att.selection.column.name,
selected.shape.names=
spatial.extent.list$selected.shape.names)
region.index <- (ttt.index & region.index)
}
} # if (!is.null(spatial.extent.list)){
# ------------------
# Subset SRD
# ------------------
srd.data <-NULL
srd.data$D.pred <- pred.data$D.pred[region.index, ]
srd.data$locs <- pred.data$locs[region.index, ]
#dim(pred.data$D.pred)
#dim(pred.data$locs)
rm(pred.data)
## KFW gc()
# ------------------
# Subset SRD
# DMF 3.17.10
# ------------------
if (!is.null(srd.max.sample.size)) {
if (nrow(srd.data$D.pred) > srd.max.sample.size){
ttt.index <- sample(x=c(1:nrow(srd.data$D.pred)),
size=srd.max.sample.size)
srd.data$D.pred <- srd.data$D.pred[ttt.index, ]
srd.data$locs <- srd.data$locs[ttt.index, ]
}
}
# --------------------------
#-----------------------------------------------------------------------------
# Load ERD parameter data & Data
#
# Wish List for erd.data.creation
# ** combine data & parameters into a single list object
# ** save only index to identify train & test sets. There is no need to
# save entire thing twice.
#------------------------------------------------------------------------------
#erdParameterListFile <- paste(data.dir, erd.data.tag, "erd.data.par.list.RData", sep="")
#load(file=erdParameterListFile) #return.list
#D.erd.data.par.list <- return.list
#erdDataFile <- paste(data.dir, D.erd.data.par.list$erd.data.tag, "erd.data.RData", sep="")
load(file=erd.data.filename) # erdDataFile) #D
# ----------------------------------------------------------------------
# Limit Spatial Extent by Point in Rectangle, Polygon, or ShapeFile
# ----------------------------------------------------------------------
xxx <- D$LONGITUDE
yyy <- D$LATITUDE
region.index <- rep(TRUE, nrow(D) )
if (!is.null(spatial.extent.list)){
## KFW if (spatial.extent.list$type == "rectangle"){
if (spatial.extent.list[['type']] == "rectangle"){
ttt.index <- (
yyy > spatial.extent.list$lat.min &
yyy < spatial.extent.list$lat.max &
xxx > spatial.extent.list$lon.min &
xxx < spatial.extent.list$lon.max )
region.index <- (ttt.index & region.index)
}
## KFW if (spatial.extent.list$type == "polygon"){
if (spatial.extent.list[['type']] == "polygon"){
ttt.index <- point.in.polygon(
xxx = xxx,
yyy = yyy,
polygon.vertices =
spatial.extent.list$polygon.vertices)
region.index <- (ttt.index & region.index)
}
## KFW if (spatial.extent.list$type == "shapefile"){
if (spatial.extent.list[['type']] == "shapefile"){
ttt.index <- point.in.shapefile(
sites = data.frame(lon=xxx, lat=yyy),
shape.dir=spatial.extent.list$shape.dir,
shape.filename=spatial.extent.list$shape.filename,
att.selection.column.name=
spatial.extent.list$att.selection.column.name,
selected.shape.names=
spatial.extent.list$selected.shape.names)
region.index <- (ttt.index & region.index)
}
} # if (!is.null(spatial.extent.list)){
# -------------------------------------------------------------------
# -------------------------------------------------------------------
# Limit Temporal Extent
# -------------------------------------------------------------------
season.index <- rep(TRUE, nrow(D) )
if (!is.null(temporal.extent.list)){
# Convert JDATE + YEAR to single temporal index
begin.tindex <- temporal.extent.list$begin.year +
temporal.extent.list$begin.jdate/366
end.tindex <- temporal.extent.list$end.year +
temporal.extent.list$end.jdate/366
D.tindex <- D$YEAR + D$DAY/366
season.index <- D.tindex >= begin.tindex &
D.tindex <= end.tindex
}
# -------------------------------------------------------------------
# ST extent
# -------------------------------------------------------------------
regional.seasonal.index <- region.index & season.index
sum(regional.seasonal.index)
ebirdReferenceDataFrame <- D[regional.seasonal.index, ]
rm(D)
## KFW gc(D)
#------------------------------------------------------------------------------------------------------------
# Train:Test Set Split : maximum coverage, eq wt train-test split
#------------------------------------------------------------------------------------------------------------
D.split <- maximum.coverage.selection(locs=data.frame(x=ebirdReferenceDataFrame$LONGITUDE ,
y=ebirdReferenceDataFrame$LATITUDE),
jdates=ebirdReferenceDataFrame$DAY,
# vector length n
grid.cell.min.lat = split.par.list$grid.cell.min.lat,
grid.cell.min.lon = split.par.list$grid.cell.min.lon,
min.val.cell.locs = split.par.list$min.val.cell.locs,
fraction.training.data = split.par.list$fraction.training.data,
mfrac = split.par.list$mfrac,
plot.it=split.par.list$plot.it)
# -----------------------------------------------
train.index <- D.split$train.index
test.index <- D.split$test.index
rm(D.split)
## KFW response.index <- names(ebirdReferenceDataFrame) %in% speciesList[[1]][1]
response.index <- names(ebirdReferenceDataFrame) %in% species.name
predictors.index <- names(ebirdReferenceDataFrame) %in% predictor.names
train.data <- NULL
test.data <- NULL
# ---------------------
train.data$X <- ebirdReferenceDataFrame[train.index, predictors.index]
ttt.y <- ebirdReferenceDataFrame[, response.index]
train.data$y <- ttt.y[train.index]
rm(ttt.y)
train.data$locs$x <- ebirdReferenceDataFrame$LONGITUDE[train.index]
train.data$locs$y <- ebirdReferenceDataFrame$LATITUDE[train.index]
train.data$jdates <- ebirdReferenceDataFrame$DAY[train.index]
# ------------------------------------
test.data$X <- ebirdReferenceDataFrame[test.index,predictors.index]
ttt.y <- ebirdReferenceDataFrame[, response.index]
test.data$y <- ttt.y[test.index]
rm(ttt.y)
test.data$locs$x <- ebirdReferenceDataFrame$LONGITUDE[test.index]
test.data$locs$y <- ebirdReferenceDataFrame$LATITUDE[test.index]
test.data$jdates <- ebirdReferenceDataFrame$DAY[test.index]
# --------------------------------------------
rm(ebirdReferenceDataFrame)
# ----------------------------------------------------------------------
# ERD parameters
# ----------------------------------------------------------------------
#protocol.tag <- D.erd.data.par.list$protocol.tag
#c("traveling","stationary") #,"casual.obs")
#if (D.erd.data.par.list$response.type == "occurrence") {
# resp.family <- "bernoulli" # "gaussian" or "bernoulli"
# }
#if (D.erd.data.par.list$response.type == "abundance"){
# resp.family <- "gaussian" # "gaussian" or "bernoulli"
# }
# --------------------------------
if ( response.family =="bernoulli") {
# ------------------------------------
# Logical - where the predicted response is the
# probability of a positive/True class
# -------------------------------------
train.data$y <- as.factor( train.data$y > 0) # TRUE if positive
test.data$y <- as.factor( test.data$y > 0) # TRUE if positive
}
# Create Truncated Gaussian Response
if ( response.family == "gaussian"){
#
# truncate responses @ 10
#sum(train.data$y > response.truncation.cutoff)
train.data$y[train.data$y > response.truncation.cutoff] <- response.truncation.cutoff
test.data$y[test.data$y >response.truncation.cutoff] <- response.truncation.cutoff
}
# --------------------------------------------------------------------------
return.list <- list(
train.index = train.index,
test.index = test.index,
#protocol.tag=protocol.tag,
response.family=response.family,
train.data = train.data,
test.data = test.data,
srd.data = srd.data)
# --------------------------------------------------------------------------
return(return.list)
}
# --------------------------------------------------------------------------
|
