forked from WinVector/RcppDynProg
-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathRcppExports.R
More file actions
414 lines (396 loc) · 13.8 KB
/
RcppExports.R
File metadata and controls
414 lines (396 loc) · 13.8 KB
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
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
# Generated by using Rcpp::compileAttributes() -> do not edit by hand
# Generator token: 10BE3573-1514-4C36-9D1C-5A225CD40393
#' const_cost
#'
#' Calculate out of sample total square error cost of using mean of points to estimate other points in interval.
#' Zero indexed.
#'
#' @param y NumericVector, values to group in order.
#' @param w NumericVector, weights.
#' @param min_seg positive integer, minimum segment size (>=1).
#' @param i integer, first index (inclusive).
#' @param j integer, j>=i last index (inclusive);
#' @return scalar, const cost of [i,...,j] interval (inclusive).
#'
#' @keywords internal
#'
#' @examples
#'
#' const_cost(c(1, 1, 2, 2), c(1, 1, 1, 1), 1, 0, 3)
#'
#' @export
const_cost <- function(y, w, min_seg, i, j) {
.Call('_RcppDynProg_const_cost', PACKAGE = 'RcppDynProg', y, w, min_seg, i, j)
}
#' const_costs
#'
#' Built matrix of total out of sample interval square error costs for held-out means.
#' One indexed.
#'
#' @param y NumericVector, values to group in order.
#' @param w NumericVector, weights.
#' @param min_seg positive integer, minimum segment size (>=1).
#' @param indices IntegerVector, order list of indices to pair.
#' @return xcosts NumericMatix, for j>=i xcosts(i,j) is the cost of partition element [i,...,j] (inclusive).
#'
#'
#' @examples
#'
#' const_costs(c(1, 1, 2, 2), c(1, 1, 1, 1), 1, 1:4)
#'
#' @export
const_costs <- function(y, w, min_seg, indices) {
.Call('_RcppDynProg_const_costs', PACKAGE = 'RcppDynProg', y, w, min_seg, indices)
}
#' const_cost_logistic
#'
#' Calculate logistic cost of using mean of points to estimate other points in interval.
#' Zero indexed.
#'
#' @param y NumericVector, 0/1 values to group in order (should be in interval [0,1]).
#' @param w NumericVector, weights (should be positive).
#' @param min_seg positive integer, minimum segment size (>=1).
#' @param i integer, first index (inclusive).
#' @param j integer, j>=i last index (inclusive);
#' @return scalar, const cost of [i,...,j] interval (inclusive).
#'
#' @keywords internal
#'
#' @examples
#'
#' const_cost_logistic(c(0.1, 0.1, 0.2, 0.2), c(1, 1, 1, 1), 1, 0, 3)
#'
#' @export
const_cost_logistic <- function(y, w, min_seg, i, j) {
.Call('_RcppDynProg_const_cost_logistic', PACKAGE = 'RcppDynProg', y, w, min_seg, i, j)
}
#' const_costs_logistic
#'
#' Built matrix of interval logistic costs for held-out means.
#' One indexed.
#'
#' @param y NumericVector, 0/1 values to group in order (should be in interval [0,1]).
#' @param w NumericVector, weights (should be positive).
#' @param min_seg positive integer, minimum segment size (>=1).
#' @param indices IntegerVector, order list of indices to pair.
#' @return xcosts NumericMatix, for j>=i xcosts(i,j) is the cost of partition element [i,...,j] (inclusive).
#'
#'
#' @examples
#'
#' const_costs_logistic(c(0.1, 0.1, 0.2, 0.2), c(1, 1, 1, 1), 1, 1:4)
#'
#' @export
const_costs_logistic <- function(y, w, min_seg, indices) {
.Call('_RcppDynProg_const_costs_logistic', PACKAGE = 'RcppDynProg', y, w, min_seg, indices)
}
#' Summarize data (for debugging).
#'
#' @param x NumericVector, expanatory variable.
#' @param y NumericVector, 0/1 values to fit.
#' @param w NumericVector, weights (required, positive).
#' @param i integer, first index (inclusive).
#' @param j integer, last index (inclusive).
#' @param skip integer, index to skip (-1 to not skip).
#' @return summary list
#'
#' @keywords internal
#'
#'
#' @examples
#'
#' costs <- matrix(c(1.5, NA ,NA ,1 ,0 , NA, 5, -1, 1), nrow = 3)
#' solve_interval_partition(costs, nrow(costs))
#'
#' @export
summarize_input <- function(x, y, w, i, j, skip) {
.Call('_RcppDynProg_summarize_input', PACKAGE = 'RcppDynProg', x, y, w, i, j, skip)
}
#' lin_cost
#'
#' Calculate cost of using linear model fit on points to estimate other points in the interval.
#' Zero indexed.
#'
#' @param x NumericVector, x-coords of values to group.
#' @param y NumericVector, values to group in order.
#' @param w NumericVector, weights.
#' @param min_seg positive integer, minimum segment size (>=1).
#' @param i integer, first index (inclusive).
#' @param j integer, j>=i last index (inclusive);
#' @return scalar, linear cost of [i,...,j] interval (inclusive).
#'
#' @keywords internal
#'
#' @examples
#'
#' lin_cost(c(1, 2, 3, 4), c(1, 2, 2, 1), c(1, 1, 1, 1), 1, 0, 3)
#'
#' @export
lin_cost <- function(x, y, w, min_seg, i, j) {
.Call('_RcppDynProg_lin_cost', PACKAGE = 'RcppDynProg', x, y, w, min_seg, i, j)
}
#' lin_costs
#'
#' Built matrix of interval costs for held-out linear models.
#' One indexed.
#'
#' @param x NumericVector, x-coords of values to group.
#' @param y NumericVector, values to group in order.
#' @param w NumericVector, weights.
#' @param min_seg positive integer, minimum segment size (>=1).
#' @param indices IntegerVector, ordered list of indices to pair.
#' @return xcosts NumericMatix, for j>=i xcosts(i,j) is the cost of partition element [i,...,j] (inclusive).
#'
#' @examples
#'
#' lin_costs(c(1, 2, 3, 4), c(1, 2, 2, 1), c(1, 1, 1, 1), 1, 1:4)
#'
#' @export
lin_costs <- function(x, y, w, min_seg, indices) {
.Call('_RcppDynProg_lin_costs', PACKAGE = 'RcppDynProg', x, y, w, min_seg, indices)
}
#' lin_cost_logistic logistic deviance pricing
#'
#' Calculate deviance cost of using logistic model fit on points to estimate other points in the interval.
#' Fits are evaluated in-sample.
#' Zero indexed.
#'
#'
#' @param x NumericVector, x-coords of values to group.
#' @param y NumericVector, values to group in order (should be in interval [0,1]).
#' @param w NumericVector, weights (positive).
#' @param min_seg positive integer, minimum segment size (>=1).
#' @param i integer, first index (inclusive).
#' @param j integer, j>=i last index (inclusive);
#' @return scalar, linear cost of [i,...,j] interval (inclusive).
#'
#' @keywords internal
#'
#' @examples
#'
#' lin_cost_logistic(c(1, 2, 3, 4, 5, 6, 7), c(0, 0, 1, 0, 1, 1, 0), c(1, 1, 1, 1, 1, 1, 1), 3, 0, 6)
#'
#' @export
lin_cost_logistic <- function(x, y, w, min_seg, i, j) {
.Call('_RcppDynProg_lin_cost_logistic', PACKAGE = 'RcppDynProg', x, y, w, min_seg, i, j)
}
#' lin_costs_logistic deviance costs.
#'
#' Built matrix of interval deviance costs for held-out logistic models.
#' Fits are evaluated in-sample.
#' One indexed.
#'
#'
#' @param x NumericVector, x-coords of values to group.
#' @param y NumericVector, values to group in order (should be in interval [0,1]).
#' @param w NumericVector, weights (should be positive).
#' @param min_seg positive integer, minimum segment size (>=1).
#' @param indices IntegerVector, ordered list of indices to pair.
#' @return xcosts NumericMatix, for j>=i xcosts(i,j) is the cost of partition element [i,...,j] (inclusive).
#'
#' @examples
#'
#' lin_costs_logistic(c(1, 2, 3, 4, 5, 6, 7), c(0, 0, 1, 0, 1, 1, 0), c(1, 1, 1, 1, 1, 1, 1), 3, 1:7)
#'
#' @export
lin_costs_logistic <- function(x, y, w, min_seg, indices) {
.Call('_RcppDynProg_lin_costs_logistic', PACKAGE = 'RcppDynProg', x, y, w, min_seg, indices)
}
#' logistic_fit
#'
#' Calculate y ~ sigmoid(a + b x) using iteratively re-weighted least squares.
#' Zero indexed.
#'
#' @param x NumericVector, expanatory variable.
#' @param y NumericVector, 0/1 values to fit.
#' @param w NumericVector, weights (required, positive).
#' @param initial_link, initial link estimates (required, all zeroes is a good start).
#' @param i integer, first index (inclusive).
#' @param j integer, last index (inclusive).
#' @param skip integer, index to skip (-1 to not skip).
#' @return vector of a and b.
#'
#' @keywords internal
#'
#' @examples
#'
#' set.seed(5)
#' d <- data.frame(
#' x = rnorm(10),
#' y = sample(c(0,1), 10, replace = TRUE)
#' )
#' weights <- runif(nrow(d))
#' m <- glm(y~x, data = d, family = binomial, weights = weights)
#' coef(m)
#' logistic_solve1(d$x, d$y, weights, rep(0.0, nrow(d)), 0, nrow(d)-1, -1)
#'
#' @export
logistic_solve1 <- function(x, y, w, initial_link, i, j, skip) {
.Call('_RcppDynProg_logistic_solve1', PACKAGE = 'RcppDynProg', x, y, w, initial_link, i, j, skip)
}
#' Out of sample logistic predictions (in link space).
#'
#' 1-hold out logistic regression predections.
#' Zero indexed.
#'
#' @param x NumericVector, expanatory variable.
#' @param y NumericVector, 0/1 values to fit.
#' @param w NumericVector, weights (required, positive).
#' @param i integer, first index (inclusive).
#' @param j integer, last index (inclusive).
#' @return vector of predictions for interval.
#'
#' @keywords internal
#'
#' @examples
#'
#' set.seed(5)
#' d <- data.frame(x = rnorm(10))
#' d$y <- d$x + rnorm(nrow(d))>0
#' weights <- runif(nrow(d))
#' m <- glm(y~x, data = d, family = binomial, weights = weights)
#' d$pred1 <- predict(m, newdata = d, type = "link")
#' d$pred2 <- xlogistic_fits(d$x, d$y, weights, 0, nrow(d)-1)
#' d <- d[order(d$x), , drop = FALSE]
#' print(d)
#'
#' @export
xlogistic_fits <- function(x, y, w, i, j) {
.Call('_RcppDynProg_xlogistic_fits', PACKAGE = 'RcppDynProg', x, y, w, i, j)
}
#' In sample logistic predictions (in link space).
#'
#' logistic regression predictions.
#' Zero indexed.
#'
#' @param x NumericVector, expanatory variable.
#' @param y NumericVector, 0/1 values to fit.
#' @param w NumericVector, weights (required, positive).
#' @param i integer, first index (inclusive).
#' @param j integer, last index (inclusive).
#' @return vector of predictions for interval.
#'
#' @keywords internal
#'
#' @examples
#'
#' set.seed(5)
#' d <- data.frame(x = rnorm(10))
#' d$y <- d$x + rnorm(nrow(d))>0
#' weights <- runif(nrow(d))
#' m <- glm(y~x, data = d, family = binomial, weights = weights)
#' d$pred1 <- predict(m, newdata = d, type = "link")
#' d$pred2 <- logistic_fits(d$x, d$y, weights, 0, nrow(d)-1)
#' d <- d[order(d$x), , drop = FALSE]
#' print(d)
#'
#' @export
logistic_fits <- function(x, y, w, i, j) {
.Call('_RcppDynProg_logistic_fits', PACKAGE = 'RcppDynProg', x, y, w, i, j)
}
#' solve_interval_partition interval partition problem with a bound on number of steps.
#'
#' Solve a for a minimal cost partition of the integers [1,...,nrow(x)] problem where for j>=i x(i,j).
#' is the cost of choosing the partition element [i,...,j].
#' Returned solution is an ordered vector v of length k<=kmax where: v[1]==1, v[k]==nrow(x)+1, and the
#' partition is of the form [v[i], v[i+1]) (intervals open on the right).
#'
#' @param x square NumericMatix, for j>=i x(i,j) is the cost of partition element [i,...,j] (inclusive).
#' @param kmax int, maximum number of segments in solution.
#' @return dynamic program solution.
#'
#' @examples
#'
#' costs <- matrix(c(1.5, NA ,NA ,1 ,0 , NA, 5, -1, 1), nrow = 3)
#' solve_interval_partition(costs, nrow(costs))
#'
#' @export
solve_interval_partition_k <- function(x, kmax) {
.Call('_RcppDynProg_solve_interval_partition_k', PACKAGE = 'RcppDynProg', x, kmax)
}
#' solve_interval_partition interval partition problem, no boun on the number of steps.
#'
#' Not working yet.
#'
#' Solve a for a minimal cost partition of the integers [1,...,nrow(x)] problem where for j>=i x(i,j).
#' is the cost of choosing the partition element [i,...,j].
#' Returned solution is an ordered vector v of length k where: v[1]==1, v[k]==nrow(x)+1, and the
#' partition is of the form [v[i], v[i+1]) (intervals open on the right).
#'
#' @param x square NumericMatix, for j>=i x(i,j) is the cost of partition element [i,...,j] (inclusive).
#' @return dynamic program solution.
#'
#' @examples
#'
#' costs <- matrix(c(1.5, NA ,NA ,1 ,0 , NA, 5, -1, 1), nrow = 3)
#' solve_interval_partition(costs, nrow(costs))
#'
#' @export
solve_interval_partition_no_k <- function(x) {
.Call('_RcppDynProg_solve_interval_partition_no_k', PACKAGE = 'RcppDynProg', x)
}
#' solve_interval_partition interval partition problem.
#'
#' Solve a for a minimal cost partition of the integers [1,...,nrow(x)] problem where for j>=i x(i,j).
#' is the cost of choosing the partition element [i,...,j].
#' Returned solution is an ordered vector v of length k<=kmax where: v[1]==1, v[k]==nrow(x)+1, and the
#' partition is of the form [v[i], v[i+1]) (intervals open on the right).
#'
#' @param x square NumericMatix, for j>=i x(i,j) is the cost of partition element [i,...,j] (inclusive).
#' @param kmax int, maximum number of segments in solution.
#' @return dynamic program solution.
#'
#' @examples
#'
#' costs <- matrix(c(1.5, NA ,NA ,1 ,0 , NA, 5, -1, 1), nrow = 3)
#' solve_interval_partition(costs, nrow(costs))
#'
#' @export
solve_interval_partition <- function(x, kmax) {
.Call('_RcppDynProg_solve_interval_partition', PACKAGE = 'RcppDynProg', x, kmax)
}
#' xlin_fits
#'
#' Calculate out of sample linear fit predictions using regularization.
#' Zero indexed.
#'
#' @param x NumericVector, explanatory variable (length>=2).
#' @param y NumericVector, values fit.
#' @param w NumericVector, weights (positive).
#' @param i integer, first index (inclusive).
#' @param j integer, j>=i+2 last index (inclusive);
#' @return vector of predictions.
#'
#' @keywords internal
#'
#' @examples
#'
#' xlin_fits(c(1, 2, 3, 4), c(1, 2, 2, 1), c(1, 1, 1, 1), 0, 3)
#'
#' @export
xlin_fits <- function(x, y, w, i, j) {
.Call('_RcppDynProg_xlin_fits', PACKAGE = 'RcppDynProg', x, y, w, i, j)
}
#' xlin_pfits
#'
#' Calculate out of sample linear fit predictions using pseudo-inverse.
#' Please see: \url{https://win-vector.com/2019/01/08/a-beautiful-2-by-2-matrix-identity/}.
#' Zero indexed.
#'
#' @param x NumericVector, explanatory variable (length>=2).
#' @param y NumericVector, values to fit.
#' @param w NumericVector, weights (positive).
#' @param i integer, first index (inclusive).
#' @param j integer, j>=i+2 last index (inclusive);
#' @return vector of predictions.
#'
#' @keywords internal
#'
#' @examples
#'
#' xlin_pfits(c(1, 2, 3, 4), c(1, 2, 2, 1), c(1, 1, 1, 1), 0, 3)
#'
#' @export
xlin_pfits <- function(x, y, w, i, j) {
.Call('_RcppDynProg_xlin_pfits', PACKAGE = 'RcppDynProg', x, y, w, i, j)
}