|
| 1 | +#------------------------------------------------------------------------------- |
| 2 | +# Copyright (c) 2019-2020 University of Newcastle upon Tyne. All rights reserved. |
| 3 | +# |
| 4 | +# This program and the accompanying materials |
| 5 | +# are made available under the terms of the GNU Public License v3.0. |
| 6 | +# |
| 7 | +# You should have received a copy of the GNU General Public License |
| 8 | +# along with this program. If not, see <http://www.gnu.org/licenses/>. |
| 9 | +#------------------------------------------------------------------------------- |
| 10 | + |
| 11 | +# |
| 12 | +# Set up |
| 13 | +# |
| 14 | + |
| 15 | +context("ds.glmPredict::smk::setup") |
| 16 | + |
| 17 | +connect.studies.dataset.cnsim(list("LAB_TSC", "LAB_TRIG", "DIS_AMI", "DIS_DIAB", "GENDER")) |
| 18 | + |
| 19 | +test_that("setup", { |
| 20 | + ds_expect_variables(c("D")) |
| 21 | +}) |
| 22 | + |
| 23 | +# |
| 24 | +# Tests |
| 25 | +# |
| 26 | + |
| 27 | +context("ds.glmPredict::smk::gaussian") |
| 28 | +test_that("simple glmPredict, gaussian, without newobj", { |
| 29 | + glmSLMA.res <- ds.glmSLMA('D$LAB_TSC~D$LAB_TRIG', family="gaussian", newobj="gaussian.glmslma.obj") |
| 30 | + |
| 31 | + expect_length(glmSLMA.res, 9) |
| 32 | + expect_equal(glmSLMA.res$num.valid.studies, 3) |
| 33 | + expect_length(glmSLMA.res$validity.check, 1) |
| 34 | + expect_equal(glmSLMA.res$validity.check, "<gaussian.glmslma.obj> appears valid in all sources") |
| 35 | + |
| 36 | + res <- ds.glmPredict("gaussian.glmslma.obj", newdataname = NULL, output.type = "response", na.action = "na.pass", newobj="gaussian.glmslma.predict.obj") |
| 37 | + |
| 38 | + expect_length(res, 3) |
| 39 | + expect_equal(class(res), "list") |
| 40 | + |
| 41 | + expect_length(res$sim1, 1) |
| 42 | + expect_length(res$sim1$safe.list, 10) |
| 43 | + expect_equal(class(res$sim1$safe.list), "list") |
| 44 | + expect_equal(res$sim1$safe.list$glm.object, "gaussian.glmslma.obj") |
| 45 | + expect_true(is.null(res$sim1$safe.list$newdfname)) |
| 46 | + expect_equal(res$sim1$safe.list$output.type, "response") |
| 47 | + expect_true(is.null(res$sim1$safe.list$dispersion)) |
| 48 | + expect_equal(res$sim1$safe.list$fit.Ntotal, 1801) |
| 49 | + expect_equal(res$sim1$safe.list$fit.Nvalid, 1801) |
| 50 | + expect_equal(res$sim1$safe.list$fit.Nmiss, 0) |
| 51 | + expect_equal(res$sim1$safe.list$fit.mean, 5.872024, tolerance = 1e-7) |
| 52 | + expect_equal(res$sim1$safe.list$fit.sd, 0.3719756, tolerance = 1e-7) |
| 53 | + expect_length(res$sim1$safe.list$fit.quantiles, 7) |
| 54 | + expect_equal(class(res$sim1$safe.list$fit.quantiles), "numeric") |
| 55 | + expect_equal(res$sim1$safe.list$fit.quantiles[[1]], 5.656127, tolerance = 1e-7) |
| 56 | + expect_equal(res$sim1$safe.list$fit.quantiles[[2]], 5.703222, tolerance = 1e-7) |
| 57 | + expect_equal(res$sim1$safe.list$fit.quantiles[[3]], 5.778930, tolerance = 1e-7) |
| 58 | + expect_equal(res$sim1$safe.list$fit.quantiles[[4]], 5.873137, tolerance = 1e-7) |
| 59 | + expect_equal(res$sim1$safe.list$fit.quantiles[[5]], 5.961123, tolerance = 1e-7) |
| 60 | + expect_equal(res$sim1$safe.list$fit.quantiles[[6]], 6.044376, tolerance = 1e-7) |
| 61 | + expect_equal(res$sim1$safe.list$fit.quantiles[[7]], 6.091611, tolerance = 1e-7) |
| 62 | + |
| 63 | + expect_length(res$sim2, 1) |
| 64 | + expect_length(res$sim2$safe.list, 10) |
| 65 | + expect_equal(class(res$sim2$safe.list), "list") |
| 66 | + expect_equal(res$sim2$safe.list$glm.object, "gaussian.glmslma.obj") |
| 67 | + expect_true(is.null(res$sim2$safe.list$newdfname)) |
| 68 | + expect_equal(res$sim2$safe.list$output.type, "response") |
| 69 | + expect_true(is.null(res$sim2$safe.list$dispersion)) |
| 70 | + expect_equal(res$sim2$safe.list$fit.Ntotal, 2526) |
| 71 | + expect_equal(res$sim2$safe.list$fit.Nvalid, 2526) |
| 72 | + expect_equal(res$sim2$safe.list$fit.Nmiss, 0) |
| 73 | + expect_equal(res$sim2$safe.list$fit.mean, 5.843564, tolerance = 1e-7) |
| 74 | + expect_equal(res$sim2$safe.list$fit.sd, 0.3027628, tolerance = 1e-7) |
| 75 | + expect_length(res$sim2$safe.list$fit.quantiles, 7) |
| 76 | + expect_equal(class(res$sim2$safe.list$fit.quantiles), "numeric") |
| 77 | + expect_equal(res$sim2$safe.list$fit.quantiles[[1]], 5.692873, tolerance = 1e-7) |
| 78 | + expect_equal(res$sim2$safe.list$fit.quantiles[[2]], 5.725140, tolerance = 1e-7) |
| 79 | + expect_equal(res$sim2$safe.list$fit.quantiles[[3]], 5.782866, tolerance = 1e-7) |
| 80 | + expect_equal(res$sim2$safe.list$fit.quantiles[[4]], 5.843818, tolerance = 1e-7) |
| 81 | + expect_equal(res$sim2$safe.list$fit.quantiles[[5]], 5.902650, tolerance = 1e-7) |
| 82 | + expect_equal(res$sim2$safe.list$fit.quantiles[[6]], 5.958954, tolerance = 1e-7) |
| 83 | + expect_equal(res$sim2$safe.list$fit.quantiles[[7]], 5.990324, tolerance = 1e-7) |
| 84 | + |
| 85 | + expect_length(res$sim3, 1) |
| 86 | + expect_length(res$sim3$safe.list, 10) |
| 87 | + expect_equal(class(res$sim3$safe.list), "list") |
| 88 | + expect_equal(res$sim3$safe.list$glm.object, "gaussian.glmslma.obj") |
| 89 | + expect_true(is.null(res$sim3$safe.list$newdfname)) |
| 90 | + expect_equal(res$sim3$safe.list$output.type, "response") |
| 91 | + expect_true(is.null(res$sim3$safe.list$dispersion)) |
| 92 | + expect_equal(res$sim3$safe.list$fit.Ntotal, 3473) |
| 93 | + expect_equal(res$sim3$safe.list$fit.Nvalid, 3473) |
| 94 | + expect_equal(res$sim3$safe.list$fit.Nmiss, 0) |
| 95 | + expect_equal(res$sim3$safe.list$fit.mean, 5.846405, tolerance = 1e-7) |
| 96 | + expect_equal(res$sim3$safe.list$fit.sd, 0.3546622, tolerance = 1e-7) |
| 97 | + expect_length(res$sim3$safe.list$fit.quantiles, 7) |
| 98 | + expect_equal(class(res$sim3$safe.list$fit.quantiles), "numeric") |
| 99 | + expect_equal(res$sim3$safe.list$fit.quantiles[[1]], 5.639923, tolerance = 1e-7) |
| 100 | + expect_equal(res$sim3$safe.list$fit.quantiles[[2]], 5.692488, tolerance = 1e-7) |
| 101 | + expect_equal(res$sim3$safe.list$fit.quantiles[[3]], 5.763677, tolerance = 1e-7) |
| 102 | + expect_equal(res$sim3$safe.list$fit.quantiles[[4]], 5.845627, tolerance = 1e-7) |
| 103 | + expect_equal(res$sim3$safe.list$fit.quantiles[[5]], 5.928798, tolerance = 1e-7) |
| 104 | + expect_equal(res$sim3$safe.list$fit.quantiles[[6]], 6.004785, tolerance = 1e-7) |
| 105 | + expect_equal(res$sim3$safe.list$fit.quantiles[[7]], 6.054574, tolerance = 1e-7) |
| 106 | +}) |
| 107 | + |
| 108 | +test_that("simple glmPredict, gaussian", { |
| 109 | + glmSLMA.res <- ds.glmSLMA('D$LAB_TSC~D$LAB_TRIG', family="gaussian", newobj="gaussian.glmslma.obj") |
| 110 | + |
| 111 | + expect_length(glmSLMA.res, 9) |
| 112 | + expect_equal(glmSLMA.res$num.valid.studies, 3) |
| 113 | + expect_length(glmSLMA.res$validity.check, 1) |
| 114 | + expect_equal(glmSLMA.res$validity.check, "<gaussian.glmslma.obj> appears valid in all sources") |
| 115 | + |
| 116 | + res <- ds.glmPredict("gaussian.glmslma.obj", output.type = "response", newobj="gaussian.glm.predict.obj") |
| 117 | + |
| 118 | + expect_length(res, 3) |
| 119 | + expect_equal(class(res), "list") |
| 120 | + |
| 121 | + expect_length(res$sim1, 1) |
| 122 | + expect_length(res$sim1$safe.list, 10) |
| 123 | + expect_equal(class(res$sim1$safe.list), "list") |
| 124 | + expect_equal(res$sim1$safe.list$glm.object, "gaussian.glmslma.obj") |
| 125 | + expect_true(is.null(res$sim1$safe.list$newdfname)) |
| 126 | + expect_equal(res$sim1$safe.list$output.type, "response") |
| 127 | + expect_true(is.null(res$sim1$safe.list$dispersion)) |
| 128 | + expect_equal(res$sim1$safe.list$fit.Ntotal, 1801) |
| 129 | + expect_equal(res$sim1$safe.list$fit.Nvalid, 1801) |
| 130 | + expect_equal(res$sim1$safe.list$fit.Nmiss, 0) |
| 131 | + expect_equal(res$sim1$safe.list$fit.mean, 5.872024, tolerance = 1e-7) |
| 132 | + expect_equal(res$sim1$safe.list$fit.sd, 0.3719756, tolerance = 1e-7) |
| 133 | + expect_length(res$sim1$safe.list$fit.quantiles, 7) |
| 134 | + expect_equal(class(res$sim1$safe.list$fit.quantiles), "numeric") |
| 135 | + expect_equal(res$sim1$safe.list$fit.quantiles[[1]], 5.656127, tolerance = 1e-7) |
| 136 | + expect_equal(res$sim1$safe.list$fit.quantiles[[2]], 5.703222, tolerance = 1e-7) |
| 137 | + expect_equal(res$sim1$safe.list$fit.quantiles[[3]], 5.778930, tolerance = 1e-7) |
| 138 | + expect_equal(res$sim1$safe.list$fit.quantiles[[4]], 5.873137, tolerance = 1e-7) |
| 139 | + expect_equal(res$sim1$safe.list$fit.quantiles[[5]], 5.961123, tolerance = 1e-7) |
| 140 | + expect_equal(res$sim1$safe.list$fit.quantiles[[6]], 6.044376, tolerance = 1e-7) |
| 141 | + expect_equal(res$sim1$safe.list$fit.quantiles[[7]], 6.091611, tolerance = 1e-7) |
| 142 | + |
| 143 | + expect_length(res$sim2, 1) |
| 144 | + expect_length(res$sim2$safe.list, 10) |
| 145 | + expect_equal(class(res$sim2$safe.list), "list") |
| 146 | + expect_equal(res$sim2$safe.list$glm.object, "gaussian.glmslma.obj") |
| 147 | + expect_true(is.null(res$sim2$safe.list$newdfname)) |
| 148 | + expect_equal(res$sim2$safe.list$output.type, "response") |
| 149 | + expect_true(is.null(res$sim2$safe.list$dispersion)) |
| 150 | + expect_equal(res$sim2$safe.list$fit.Ntotal, 2526) |
| 151 | + expect_equal(res$sim2$safe.list$fit.Nvalid, 2526) |
| 152 | + expect_equal(res$sim2$safe.list$fit.Nmiss, 0) |
| 153 | + expect_equal(res$sim2$safe.list$fit.mean, 5.843564, tolerance = 1e-7) |
| 154 | + expect_equal(res$sim2$safe.list$fit.sd, 0.3027628, tolerance = 1e-7) |
| 155 | + expect_length(res$sim2$safe.list$fit.quantiles, 7) |
| 156 | + expect_equal(class(res$sim2$safe.list$fit.quantiles), "numeric") |
| 157 | + expect_equal(res$sim2$safe.list$fit.quantiles[[1]], 5.692873, tolerance = 1e-7) |
| 158 | + expect_equal(res$sim2$safe.list$fit.quantiles[[2]], 5.725140, tolerance = 1e-7) |
| 159 | + expect_equal(res$sim2$safe.list$fit.quantiles[[3]], 5.782866, tolerance = 1e-7) |
| 160 | + expect_equal(res$sim2$safe.list$fit.quantiles[[4]], 5.843818, tolerance = 1e-7) |
| 161 | + expect_equal(res$sim2$safe.list$fit.quantiles[[5]], 5.902650, tolerance = 1e-7) |
| 162 | + expect_equal(res$sim2$safe.list$fit.quantiles[[6]], 5.958954, tolerance = 1e-7) |
| 163 | + expect_equal(res$sim2$safe.list$fit.quantiles[[7]], 5.990324, tolerance = 1e-7) |
| 164 | + |
| 165 | + expect_length(res$sim3, 1) |
| 166 | + expect_length(res$sim3$safe.list, 10) |
| 167 | + expect_equal(class(res$sim3$safe.list), "list") |
| 168 | + expect_equal(res$sim3$safe.list$glm.object, "gaussian.glmslma.obj") |
| 169 | + expect_true(is.null(res$sim3$safe.list$newdfname)) |
| 170 | + expect_equal(res$sim3$safe.list$output.type, "response") |
| 171 | + expect_true(is.null(res$sim3$safe.list$dispersion)) |
| 172 | + expect_equal(res$sim3$safe.list$fit.Ntotal, 3473) |
| 173 | + expect_equal(res$sim3$safe.list$fit.Nvalid, 3473) |
| 174 | + expect_equal(res$sim3$safe.list$fit.Nmiss, 0) |
| 175 | + expect_equal(res$sim3$safe.list$fit.mean, 5.846405, tolerance = 1e-7) |
| 176 | + expect_equal(res$sim3$safe.list$fit.sd, 0.3546622, tolerance = 1e-7) |
| 177 | + expect_length(res$sim3$safe.list$fit.quantiles, 7) |
| 178 | + expect_equal(class(res$sim3$safe.list$fit.quantiles), "numeric") |
| 179 | + expect_equal(res$sim3$safe.list$fit.quantiles[[1]], 5.639923, tolerance = 1e-7) |
| 180 | + expect_equal(res$sim3$safe.list$fit.quantiles[[2]], 5.692488, tolerance = 1e-7) |
| 181 | + expect_equal(res$sim3$safe.list$fit.quantiles[[3]], 5.763677, tolerance = 1e-7) |
| 182 | + expect_equal(res$sim3$safe.list$fit.quantiles[[4]], 5.845627, tolerance = 1e-7) |
| 183 | + expect_equal(res$sim3$safe.list$fit.quantiles[[5]], 5.928798, tolerance = 1e-7) |
| 184 | + expect_equal(res$sim3$safe.list$fit.quantiles[[6]], 6.004785, tolerance = 1e-7) |
| 185 | + expect_equal(res$sim3$safe.list$fit.quantiles[[7]], 6.054574, tolerance = 1e-7) |
| 186 | +}) |
| 187 | + |
| 188 | +context("ds.glmPredict::smk::poisson") |
| 189 | +test_that("simple glmPredict, poisson, without newobj", { |
| 190 | + glmSLMA.res <- ds.glmSLMA('D$LAB_TSC~D$LAB_TRIG', family="poisson", newobj="poisson.glmslma.obj") |
| 191 | + |
| 192 | + expect_length(glmSLMA.res, 9) |
| 193 | + expect_equal(glmSLMA.res$num.valid.studies, 3) |
| 194 | + expect_length(glmSLMA.res$validity.check, 1) |
| 195 | + expect_equal(glmSLMA.res$validity.check, "<poisson.glmslma.obj> appears valid in all sources") |
| 196 | + |
| 197 | + res <- ds.glmPredict("poisson.glmslma.obj", output.type = "response") |
| 198 | + |
| 199 | + expect_length(res, 3) |
| 200 | + expect_equal(class(res), "list") |
| 201 | + |
| 202 | + expect_length(res$sim1, 1) |
| 203 | + expect_length(res$sim1$safe.list, 10) |
| 204 | + expect_equal(class(res$sim1$safe.list), "list") |
| 205 | + expect_equal(res$sim1$safe.list$glm.object, "poisson.glmslma.obj") |
| 206 | + expect_true(is.null(res$sim1$safe.list$newdfname)) |
| 207 | + expect_equal(res$sim1$safe.list$output.type, "response") |
| 208 | + expect_true(is.null(res$sim1$safe.list$dispersion)) |
| 209 | + expect_equal(res$sim1$safe.list$fit.Ntotal, 1801) |
| 210 | + expect_equal(res$sim1$safe.list$fit.Nvalid, 1801) |
| 211 | + expect_equal(res$sim1$safe.list$fit.Nmiss, 0) |
| 212 | + expect_equal(res$sim1$safe.list$fit.mean, 5.872024, tolerance = 1e-7) |
| 213 | + expect_equal(res$sim1$safe.list$fit.sd, 0.3720242, tolerance = 1e-7) |
| 214 | + expect_length(res$sim1$safe.list$fit.quantiles, 7) |
| 215 | + expect_equal(class(res$sim1$safe.list$fit.quantiles), "numeric") |
| 216 | + expect_equal(res$sim1$safe.list$fit.quantiles[[1]], 5.659318, tolerance = 1e-7) |
| 217 | + expect_equal(res$sim1$safe.list$fit.quantiles[[2]], 5.704707, tolerance = 1e-7) |
| 218 | + expect_equal(res$sim1$safe.list$fit.quantiles[[3]], 5.778435, tolerance = 1e-7) |
| 219 | + expect_equal(res$sim1$safe.list$fit.quantiles[[4]], 5.871512, tolerance = 1e-7) |
| 220 | + expect_equal(res$sim1$safe.list$fit.quantiles[[5]], 5.959795, tolerance = 1e-7) |
| 221 | + expect_equal(res$sim1$safe.list$fit.quantiles[[6]], 6.044552, tolerance = 1e-7) |
| 222 | + expect_equal(res$sim1$safe.list$fit.quantiles[[7]], 6.093175, tolerance = 1e-7) |
| 223 | + |
| 224 | + expect_length(res$sim2, 1) |
| 225 | + expect_length(res$sim2$safe.list, 10) |
| 226 | + expect_equal(class(res$sim2$safe.list), "list") |
| 227 | + expect_equal(res$sim2$safe.list$glm.object, "poisson.glmslma.obj") |
| 228 | + expect_true(is.null(res$sim2$safe.list$newdfname)) |
| 229 | + expect_equal(res$sim2$safe.list$output.type, "response") |
| 230 | + expect_true(is.null(res$sim2$safe.list$dispersion)) |
| 231 | + expect_equal(res$sim2$safe.list$fit.Ntotal, 2526) |
| 232 | + expect_equal(res$sim2$safe.list$fit.Nvalid, 2526) |
| 233 | + expect_equal(res$sim2$safe.list$fit.Nmiss, 0) |
| 234 | + expect_equal(res$sim2$safe.list$fit.mean, 5.843564, tolerance = 1e-7) |
| 235 | + expect_equal(res$sim2$safe.list$fit.sd, 0.3027769, tolerance = 1e-7) |
| 236 | + expect_length(res$sim2$safe.list$fit.quantiles, 7) |
| 237 | + expect_equal(class(res$sim2$safe.list$fit.quantiles), "numeric") |
| 238 | + expect_equal(res$sim2$safe.list$fit.quantiles[[1]], 5.694361, tolerance = 1e-7) |
| 239 | + expect_equal(res$sim2$safe.list$fit.quantiles[[2]], 5.725835, tolerance = 1e-7) |
| 240 | + expect_equal(res$sim2$safe.list$fit.quantiles[[3]], 5.782577, tolerance = 1e-7) |
| 241 | + expect_equal(res$sim2$safe.list$fit.quantiles[[4]], 5.843101, tolerance = 1e-7) |
| 242 | + expect_equal(res$sim2$safe.list$fit.quantiles[[5]], 5.902120, tolerance = 1e-7) |
| 243 | + expect_equal(res$sim2$safe.list$fit.quantiles[[6]], 5.959162, tolerance = 1e-7) |
| 244 | + expect_equal(res$sim2$safe.list$fit.quantiles[[7]], 5.991181, tolerance = 1e-7) |
| 245 | + |
| 246 | + expect_length(res$sim3, 1) |
| 247 | + expect_length(res$sim3$safe.list, 10) |
| 248 | + expect_equal(class(res$sim3$safe.list), "list") |
| 249 | + expect_equal(res$sim3$safe.list$glm.object, "poisson.glmslma.obj") |
| 250 | + expect_true(is.null(res$sim3$safe.list$newdfname)) |
| 251 | + expect_equal(res$sim3$safe.list$output.type, "response") |
| 252 | + expect_true(is.null(res$sim3$safe.list$dispersion)) |
| 253 | + expect_equal(res$sim3$safe.list$fit.Ntotal, 3473) |
| 254 | + expect_equal(res$sim3$safe.list$fit.Nvalid, 3473) |
| 255 | + expect_equal(res$sim3$safe.list$fit.Nmiss, 0) |
| 256 | + expect_equal(res$sim3$safe.list$fit.mean, 5.846405, tolerance = 1e-7) |
| 257 | + expect_equal(res$sim3$safe.list$fit.sd, 0.3546933, tolerance = 1e-7) |
| 258 | + expect_length(res$sim3$safe.list$fit.quantiles, 7) |
| 259 | + expect_equal(class(res$sim3$safe.list$fit.quantiles), "numeric") |
| 260 | + expect_equal(res$sim3$safe.list$fit.quantiles[[1]], 5.642464, tolerance = 1e-7) |
| 261 | + expect_equal(res$sim3$safe.list$fit.quantiles[[2]], 5.693362, tolerance = 1e-7) |
| 262 | + expect_equal(res$sim3$safe.list$fit.quantiles[[3]], 5.763027, tolerance = 1e-7) |
| 263 | + expect_equal(res$sim3$safe.list$fit.quantiles[[4]], 5.844277, tolerance = 1e-7) |
| 264 | + expect_equal(res$sim3$safe.list$fit.quantiles[[5]], 5.927911, tolerance = 1e-7) |
| 265 | + expect_equal(res$sim3$safe.list$fit.quantiles[[6]], 6.005366, tolerance = 1e-7) |
| 266 | + expect_equal(res$sim3$safe.list$fit.quantiles[[7]], 6.056664, tolerance = 1e-7) |
| 267 | +}) |
| 268 | + |
| 269 | +test_that("simple glmPredict, poisson", { |
| 270 | + glmSLMA.res <- ds.glmSLMA('D$LAB_TSC~D$LAB_TRIG', family="poisson", newobj="poisson.glmslma.obj") |
| 271 | + |
| 272 | + expect_length(glmSLMA.res, 9) |
| 273 | + expect_equal(glmSLMA.res$num.valid.studies, 3) |
| 274 | + expect_length(glmSLMA.res$validity.check, 1) |
| 275 | + expect_equal(glmSLMA.res$validity.check, "<poisson.glmslma.obj> appears valid in all sources") |
| 276 | + |
| 277 | + res <- ds.glmPredict("poisson.glmslma.obj", output.type = "response", newobj="poisson.glm.predict.obj") |
| 278 | + |
| 279 | + expect_length(res, 3) |
| 280 | + expect_equal(class(res), "list") |
| 281 | + |
| 282 | + expect_length(res$sim1, 1) |
| 283 | + expect_length(res$sim1$safe.list, 10) |
| 284 | + expect_equal(class(res$sim1$safe.list), "list") |
| 285 | + expect_equal(res$sim1$safe.list$glm.object, "poisson.glmslma.obj") |
| 286 | + expect_true(is.null(res$sim1$safe.list$newdfname)) |
| 287 | + expect_equal(res$sim1$safe.list$output.type, "response") |
| 288 | + expect_true(is.null(res$sim1$safe.list$dispersion)) |
| 289 | + expect_equal(res$sim1$safe.list$fit.Ntotal, 1801) |
| 290 | + expect_equal(res$sim1$safe.list$fit.Nvalid, 1801) |
| 291 | + expect_equal(res$sim1$safe.list$fit.Nmiss, 0) |
| 292 | + expect_equal(res$sim1$safe.list$fit.mean, 5.872024, tolerance = 1e-7) |
| 293 | + expect_equal(res$sim1$safe.list$fit.sd, 0.3720242, tolerance = 1e-7) |
| 294 | + expect_length(res$sim1$safe.list$fit.quantiles, 7) |
| 295 | + expect_equal(class(res$sim1$safe.list$fit.quantiles), "numeric") |
| 296 | + expect_equal(res$sim1$safe.list$fit.quantiles[[1]], 5.659318, tolerance = 1e-7) |
| 297 | + expect_equal(res$sim1$safe.list$fit.quantiles[[2]], 5.704707, tolerance = 1e-7) |
| 298 | + expect_equal(res$sim1$safe.list$fit.quantiles[[3]], 5.778435, tolerance = 1e-7) |
| 299 | + expect_equal(res$sim1$safe.list$fit.quantiles[[4]], 5.871512, tolerance = 1e-7) |
| 300 | + expect_equal(res$sim1$safe.list$fit.quantiles[[5]], 5.959795, tolerance = 1e-7) |
| 301 | + expect_equal(res$sim1$safe.list$fit.quantiles[[6]], 6.044552, tolerance = 1e-7) |
| 302 | + expect_equal(res$sim1$safe.list$fit.quantiles[[7]], 6.093175, tolerance = 1e-7) |
| 303 | + |
| 304 | + expect_length(res$sim2, 1) |
| 305 | + expect_length(res$sim2$safe.list, 10) |
| 306 | + expect_equal(class(res$sim2$safe.list), "list") |
| 307 | + expect_equal(res$sim2$safe.list$glm.object, "poisson.glmslma.obj") |
| 308 | + expect_true(is.null(res$sim2$safe.list$newdfname)) |
| 309 | + expect_equal(res$sim2$safe.list$output.type, "response") |
| 310 | + expect_true(is.null(res$sim2$safe.list$dispersion)) |
| 311 | + expect_equal(res$sim2$safe.list$fit.Ntotal, 2526) |
| 312 | + expect_equal(res$sim2$safe.list$fit.Nvalid, 2526) |
| 313 | + expect_equal(res$sim2$safe.list$fit.Nmiss, 0) |
| 314 | + expect_equal(res$sim2$safe.list$fit.mean, 5.843564, tolerance = 1e-7) |
| 315 | + expect_equal(res$sim2$safe.list$fit.sd, 0.3027769, tolerance = 1e-7) |
| 316 | + expect_length(res$sim2$safe.list$fit.quantiles, 7) |
| 317 | + expect_equal(class(res$sim2$safe.list$fit.quantiles), "numeric") |
| 318 | + expect_equal(res$sim2$safe.list$fit.quantiles[[1]], 5.694361, tolerance = 1e-7) |
| 319 | + expect_equal(res$sim2$safe.list$fit.quantiles[[2]], 5.725835, tolerance = 1e-7) |
| 320 | + expect_equal(res$sim2$safe.list$fit.quantiles[[3]], 5.782577, tolerance = 1e-7) |
| 321 | + expect_equal(res$sim2$safe.list$fit.quantiles[[4]], 5.843101, tolerance = 1e-7) |
| 322 | + expect_equal(res$sim2$safe.list$fit.quantiles[[5]], 5.902120, tolerance = 1e-7) |
| 323 | + expect_equal(res$sim2$safe.list$fit.quantiles[[6]], 5.959162, tolerance = 1e-7) |
| 324 | + expect_equal(res$sim2$safe.list$fit.quantiles[[7]], 5.991181, tolerance = 1e-7) |
| 325 | + |
| 326 | + expect_length(res$sim3, 1) |
| 327 | + expect_length(res$sim3$safe.list, 10) |
| 328 | + expect_equal(class(res$sim3$safe.list), "list") |
| 329 | + expect_equal(res$sim3$safe.list$glm.object, "poisson.glmslma.obj") |
| 330 | + expect_true(is.null(res$sim3$safe.list$newdfname)) |
| 331 | + expect_equal(res$sim3$safe.list$output.type, "response") |
| 332 | + expect_true(is.null(res$sim3$safe.list$dispersion)) |
| 333 | + expect_equal(res$sim3$safe.list$fit.Ntotal, 3473) |
| 334 | + expect_equal(res$sim3$safe.list$fit.Nvalid, 3473) |
| 335 | + expect_equal(res$sim3$safe.list$fit.Nmiss, 0) |
| 336 | + expect_equal(res$sim3$safe.list$fit.mean, 5.846405, tolerance = 1e-7) |
| 337 | + expect_equal(res$sim3$safe.list$fit.sd, 0.3546933, tolerance = 1e-7) |
| 338 | + expect_length(res$sim3$safe.list$fit.quantiles, 7) |
| 339 | + expect_equal(class(res$sim3$safe.list$fit.quantiles), "numeric") |
| 340 | + expect_equal(res$sim3$safe.list$fit.quantiles[[1]], 5.642464, tolerance = 1e-7) |
| 341 | + expect_equal(res$sim3$safe.list$fit.quantiles[[2]], 5.693362, tolerance = 1e-7) |
| 342 | + expect_equal(res$sim3$safe.list$fit.quantiles[[3]], 5.763027, tolerance = 1e-7) |
| 343 | + expect_equal(res$sim3$safe.list$fit.quantiles[[4]], 5.844277, tolerance = 1e-7) |
| 344 | + expect_equal(res$sim3$safe.list$fit.quantiles[[5]], 5.927911, tolerance = 1e-7) |
| 345 | + expect_equal(res$sim3$safe.list$fit.quantiles[[6]], 6.005366, tolerance = 1e-7) |
| 346 | + expect_equal(res$sim3$safe.list$fit.quantiles[[7]], 6.056664, tolerance = 1e-7) |
| 347 | +}) |
| 348 | + |
| 349 | +# |
| 350 | +# Shutdown |
| 351 | +# |
| 352 | + |
| 353 | +context("ds.glmPredict::smk::shutdown") |
| 354 | + |
| 355 | +test_that("shutdown", { |
| 356 | + ds_expect_variables(c("D", "gaussian.glmslma.obj", "gaussian.glmslma.predict.obj", "gaussian.glm.predict.obj","poisson.glm.predict.obj", "poisson.glmslma.obj", "predict_glm")) |
| 357 | +}) |
| 358 | + |
| 359 | +disconnect.studies.dataset.cnsim() |
| 360 | + |
| 361 | +# |
| 362 | +# Done |
| 363 | +# |
| 364 | + |
| 365 | +context("ds.glmPredict::smk::done") |
0 commit comments