R/get.W.R In dmt: Dependency Modeling Toolkit

Defines functions get.W.nonneg.identicalget.W.nonnegget.W

```# (C) 2008-2011 Leo Lahti and Olli-Pekka Huovilainen
# FreeBSD License (keep this notice)

# "Reality is merely an illusion, albeit a very persistent one."
# - Albert Einstein

get.W <- function (vec, Dim) {

# (C) 2008-2011 Leo Lahti and Olli-Pekka Huovilainen
# FreeBSD License (keep this notice)

# Convert parameter vector into matrices Wx and Wy
W <- list()
W\$X <- array(vec[1:(Dim\$X*Dim\$Z)], dim = c(Dim\$X,Dim\$Z))
T <- array(vec[-seq(Dim\$X*Dim\$Z)], dim = c(Dim\$Y,Dim\$X))
W\$Y <- T%*%W\$X

list(W = W, T = T)
}

get.W.nonneg <- function (vec, Dim) {

# (C) 2008-2011 Leo Lahti and Olli-Pekka Huovilainen
# FreeBSD License (keep this notice)

# Convert parameter vector into matrices Wx and Wy
# in general Wx != Wy

# Note that we always assume that W is positive Therefore remove the
#sign here to speed up optimization
vec <- abs(vec)

# NOTE: assumes dx = dy
W <- list()
nx <- Dim\$X*Dim\$Z
W\$X <- array(vec[1:nx], dim = c(Dim\$X, Dim\$Z))
W\$Y <- array(vec[(nx + 1):length(vec)], dim = c(Dim\$Y, Dim\$Z))
W\$total <- rbind(W\$X, W\$Y)

W

}

get.W.nonneg.identical <- function (vec, Dim) {

# (C) 2008-2011 Leo Lahti and Olli-Pekka Huovilainen
# FreeBSD License (keep this notice)

# Convert parameter vector into matrix
# Note that we always assume that W is positive
# Therefore remove the sign here to speed up optimization

array(abs(vec), dim = c(Dim\$X, Dim\$Z))

}
```

Try the dmt package in your browser

Any scripts or data that you put into this service are public.

dmt documentation built on May 29, 2017, 11 a.m.