// NewNormal creates a new Normal with the given mean and covariance matrix.
// NewNormal panics if len(mu) == 0, or if len(mu) != sigma.N. If the covariance
// matrix is not positive-definite, the returned boolean is false.
func NewNormal(mu []float64, sigma mat64.Symmetric, src *rand.Rand) (*Normal, bool) {
if len(mu) == 0 {
panic(badZeroDimension)
}
dim := sigma.Symmetric()
if dim != len(mu) {
panic(badSizeMismatch)
}
n := &Normal{
src: src,
dim: dim,
mu: make([]float64, dim),
sigma: mat64.NewSymDense(dim, nil),
chol: mat64.NewTriDense(dim, true, nil),
}
copy(n.mu, mu)
n.sigma.CopySym(sigma)
// TODO(btracey): Change this to the input Sigma, in case it is diagonal or
// banded.
ok := n.chol.Cholesky(n.sigma, true)
if !ok {
return nil, false
}
for i := 0; i < dim; i++ {
n.logSqrtDet += math.Log(n.chol.At(i, i))
}
return n, true
}
// NewNormal creates a new Normal with the given mean and covariance matrix.
// NewNormal panics if len(mu) == 0, or if len(mu) != sigma.N. If the covariance
// matrix is not positive-definite, the returned boolean is false.
func NewNormal(mu []float64, sigma mat64.Symmetric, src *rand.Rand) (*Normal, bool) {
if len(mu) == 0 {
panic(badZeroDimension)
}
dim := sigma.Symmetric()
if dim != len(mu) {
panic(badSizeMismatch)
}
n := &Normal{
src: src,
dim: dim,
mu: make([]float64, dim),
}
copy(n.mu, mu)
ok := n.chol.Factorize(sigma)
if !ok {
return nil, false
}
n.lower.LFromCholesky(&n.chol)
n.logSqrtDet = 0.5 * n.chol.LogDet()
return n, true
}