Single-hidden-layer MLP genomic predictor (base R)

Usage

morie_deep_learning_genomic(
  x,
  y,
  markers,
  hidden = 16,
  n_epochs = 200,
  lr = 0.01,
  l2 = 0.001,
  seed = 0,
  deterministic_seed = NULL
)

Arguments

x: Fixed-effect design (optional).
y: Numeric response.
markers: Genotype matrix (n x m).
hidden: Hidden units (default 16).
n_epochs: Training epochs.
lr: Learning rate.
l2: L2 weight decay.
seed: Seed.
deterministic_seed: Optional integer; if supplied, RNG state is derived via morie_det_rng() keyed on ("dlgen", deterministic_seed) so Py<->R streams agree on the canonical fixture. When NULL (default) behaviour is unchanged.

Value

list(estimate, y_hat, beta, W1, b1, w2, b2, se, n, method).

References

Montesinos Lopez Ch 12.

Examples

morie_deep_learning_genomic(
  x = rnorm(50), y = rnorm(50),
  markers = matrix(sample(0:2, 200, TRUE), 50, 4)
)
#> $estimate
#> [1] 0.01107817
#> 
#> $y_hat
#>  [1] -0.02179151  0.07430120  0.24156595  0.29541150  0.05569298 -0.19144661
#>  [7]  0.30328142  0.30328142 -0.13194797  0.03209409 -0.04894465 -0.15787242
#> [13]  0.13001407 -0.15787242  0.01805571 -0.06172872  0.25354045 -0.04956823
#> [19]  0.05569298  0.11440458  0.29541150 -0.34395711 -0.06172872 -0.03733046
#> [25] -0.38657652  0.05569298 -0.34395711 -0.24544857 -0.06580968  0.11469453
#> [31]  0.03133289  0.02466530  0.05569298 -0.37599710  0.13638153  0.22752225
#> [37]  0.30328142  0.09991966  0.22752225 -0.24544857  0.05272527 -0.04722832
#> [43]  0.28023765 -0.07818958 -0.19144661 -0.15787242  0.24156595  0.08391140
#> [49]  0.24364732 -0.39546923
#> 
#> $beta
#> numeric(0)
#> 
#> $W1
#>             [,1]        [,2]       [,3]       [,4]       [,5]       [,6]
#> [1,] -0.09400292 -0.46284502  0.4123016 -0.5066601  0.8391846 1.24125625
#> [2,]  0.76185332 -0.08155816 -0.6020754  0.2058128  0.7875070 0.38475816
#> [3,] -0.30970018  0.21997113 -0.5204349 -0.1901563 -0.1708525 0.29969415
#> [4,] -0.17601092 -0.36001619  0.7218800  0.2045870 -1.1446730 0.02280322
#>             [,7]       [,8]        [,9]        [,10]      [,11]      [,12]
#> [1,]  0.23623592 -0.7123171 -0.62500936 -0.001120298  0.8893374  0.1292674
#> [2,] -0.03204853  0.4645252  0.33026831 -0.311098173 -0.1852676 -0.4903363
#> [3,]  0.28220426  0.6671459  0.01747503 -0.162053893 -0.8355676 -1.4376705
#> [4,] -0.16069105 -0.1548957 -0.84674134 -0.577386876  0.1234642 -0.3205078
#>            [,13]        [,14]      [,15]       [,16]
#> [1,]  0.28590475 -0.590720820  0.5033317 -1.00707641
#> [2,] -0.04280214  0.547569752  0.1351999 -0.18284897
#> [3,] -0.08385446 -0.004567451 -0.3960267 -0.04575457
#> [4,]  0.26825409  0.352653467  0.6009179 -0.03691863
#> 
#> $b1
#>  [1] -6.230404e-03  1.574187e-03 -1.397020e-03 -1.382204e-04  1.031121e-02
#>  [6] -4.189578e-02 -1.399986e-02  1.105571e-03 -1.075601e-03 -1.862857e-03
#> [11] -7.407350e-03 -4.499677e-04 -1.376838e-03  5.090840e-04  6.262251e-05
#> [16] -5.896144e-02
#> 
#> $w2
#>  [1]  0.225633936  0.063620017 -0.050768949  0.008399112 -0.078243475
#>  [6]  0.280363659  0.314069221 -0.393172992  0.063893413 -0.010473754
#> [11] -0.294811925 -0.096050832 -0.100387392  0.041095149  0.150676961
#> [16]  0.382044787
#> 
#> $b2
#> [1] 0.01836297
#> 
#> $loss_curve
#>   [1] 1.4300532 1.4016491 1.3747865 1.3493710 1.3253145 1.3025346 1.2809544
#>   [8] 1.2605020 1.2411101 1.2227159 1.2052604 1.1886885 1.1729486 1.1579924
#>  [15] 1.1437746 1.1302527 1.1173869 1.1051399 1.0934767 1.0823645 1.0717724
#>  [22] 1.0616716 1.0520348 1.0428365 1.0340529 1.0256613 1.0176406 1.0099709
#>  [29] 1.0026336 0.9956109 0.9888864 0.9824445 0.9762705 0.9703506 0.9646718
#>  [36] 0.9592219 0.9539893 0.9489631 0.9441332 0.9394897 0.9350237 0.9307265
#>  [43] 0.9265900 0.9226065 0.9187687 0.9150699 0.9115035 0.9080635 0.9047441
#>  [50] 0.9015398 0.8984454 0.8954561 0.8925672 0.8897744 0.8870734 0.8844603
#>  [57] 0.8819314 0.8794832 0.8771123 0.8748155 0.8725897 0.8704321 0.8683400
#>  [64] 0.8663108 0.8643419 0.8624310 0.8605759 0.8587745 0.8570248 0.8553247
#>  [71] 0.8536724 0.8520662 0.8505045 0.8489855 0.8475079 0.8460700 0.8446705
#>  [78] 0.8433082 0.8419816 0.8406896 0.8394311 0.8382048 0.8370097 0.8358449
#>  [85] 0.8347092 0.8336018 0.8325217 0.8314681 0.8304402 0.8294370 0.8284580
#>  [92] 0.8275022 0.8265691 0.8256579 0.8247679 0.8238985 0.8230492 0.8222192
#>  [99] 0.8214081 0.8206153 0.8198402 0.8190823 0.8183413 0.8176164 0.8169074
#> [106] 0.8162138 0.8155351 0.8148709 0.8142208 0.8135845 0.8129615 0.8123514
#> [113] 0.8117540 0.8111689 0.8105958 0.8100343 0.8094841 0.8089450 0.8084166
#> [120] 0.8078986 0.8073908 0.8068929 0.8064047 0.8059259 0.8054562 0.8049955
#> [127] 0.8045434 0.8040998 0.8036645 0.8032372 0.8028177 0.8024059 0.8020015
#> [134] 0.8016044 0.8012143 0.8008311 0.8004547 0.8000848 0.7997213 0.7993641
#> [141] 0.7990129 0.7986676 0.7983282 0.7979943 0.7976660 0.7973430 0.7970253
#> [148] 0.7967126 0.7964050 0.7961022 0.7958041 0.7955107 0.7952218 0.7949373
#> [155] 0.7946571 0.7943810 0.7941091 0.7938412 0.7935771 0.7933169 0.7930603
#> [162] 0.7928074 0.7925581 0.7923121 0.7920696 0.7918303 0.7915943 0.7913613
#> [169] 0.7911315 0.7909046 0.7906806 0.7904595 0.7902412 0.7900256 0.7898126
#> [176] 0.7896022 0.7893943 0.7891889 0.7889859 0.7887853 0.7885870 0.7883908
#> [183] 0.7881969 0.7880052 0.7878155 0.7876278 0.7874421 0.7872584 0.7870766
#> [190] 0.7868966 0.7867185 0.7865421 0.7863675 0.7861945 0.7860232 0.7858535
#> [197] 0.7856853 0.7855187 0.7853536 0.7851900
#> 
#> $se
#> [1] 0.8860179
#> 
#> $n
#> [1] 50
#> 
#> $method
#> [1] "MLP-1H base-R"
#>