close #1875

norm(A, :fro) is replaced with normfro(A).

Author: ViralBShah

base/export.jl

@@ -707,6 +707,7 @@ export
     lud,
     lud!,
     norm,
+    normfro,
     null,
     pinv,
     qr,

base/linalg.jl

@@ -33,7 +33,7 @@ diag(A::AbstractVector) = error("Perhaps you meant to use diagm().")
 
 function norm{T}(x::AbstractVector{T}, p::Number)
     if length(x) == 0
-        a = zero(eltype(x))
+        a = zero(T)
     elseif p == Inf
         a = max(abs(x))
     elseif p == -Inf
@@ -53,7 +53,7 @@ end
 norm{T<:Integer}(x::AbstractVector{T}, p::Number) = norm(float(x), p)
 norm(x::AbstractVector) = norm(x, 2)
 
-function norm(A::AbstractMatrix, p)
+function norm(A::AbstractMatrix, p::Number)
     m, n = size(A)
     if m == 0 || n == 0
         a = zero(eltype(A))
@@ -65,10 +65,8 @@ function norm(A::AbstractMatrix, p)
         a = max(svdvals(A))
     elseif p == Inf
         a = max(sum(abs(A),2))
-    elseif p == :fro
-        a = norm(reshape(A, length(A)))
     else
-        error("invalid parameter to matrix norm")
+        error("invalid parameter p given to compute matrix norm")
     end
     return float(a)
 end
@@ -78,6 +76,9 @@ norm(A::AbstractMatrix) = norm(A, 2)
 norm(x::Number) = abs(x)
 norm(x::Number, p) = abs(x)
 
+normfro(A::AbstractMatrix) = norm(reshape(A, length(A)), 2)
+normfro(x::Number) = abs(x)
+
 rank(A::AbstractMatrix, tol::Real) = sum(svdvals(A) .> tol)
 function rank(A::AbstractMatrix)
     m,n = size(A)

base/linalg_dense.jl

@@ -779,7 +779,7 @@ end
 svd(A) = svd(A,true,false)
 svd(A, thin::Bool) = svd(A,true,thin)
 
-svdvals(A) = svd(A,false,true)[2]
+svdvals(A) = svdt(A,false,true)[2]
 
 
 function (\){T<:BlasFloat}(A::StridedMatrix{T}, B::StridedVecOrMat{T})

doc/stdlib/base.rst

@@ -1761,11 +1761,15 @@ Linear algebra functions in Julia are largely implemented by calling functions f
 
 .. function:: norm(A, [p])
 
-   Compute the p-norm of a vector or a matrix. ``p`` is ``2`` by default, if not provided. If ``A`` is a matrix, valid values for ``p`` are ``1``, ``2``, ``Inf``, or ``:fro`` (Frobenius norm).
+   Compute the ``p``-norm of a vector or a matrix. ``p`` is ``2`` by default, if not provided. If ``A`` is a vector, ``norm(A, p)`` computes the ``p``-norm. ``norm(A, Inf)`` returns the largest value in ``abs(A)``, whereas ``norm(A, -Inf)`` returns the smallest. If ``A`` is a matrix, valid values for ``p`` are ``1``, ``2``, or ``Inf``. In order to compute the Frobenius norm, use ``normfro``.
+
+.. function:: normfro(A)
+
+   Compute the Frobenius norm of a matrix ``A``.
 
 .. function:: cond(M, [p])
 
-   Matrix condition number, computed using the p-norm. ``p`` is 2 by default, if not provided. Valid values for ``p`` are ``1``, ``2``, ``Inf``, or ``:fro`` (Frobenius norm).
+   Matrix condition number, computed using the p-norm. ``p`` is 2 by default, if not provided. Valid values for ``p`` are ``1``, ``2``, or ``Inf``.
 
 .. function:: trace(M)