We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
1 parent 26fac4b commit 455e6a6Copy full SHA for 455e6a6
Extras/Inverse eigenvalue problem.jl
@@ -12,14 +12,14 @@ import ApproxFun: bandwidth
12
# To answer this question, we use a dual number to observe the sensitivity in λ₀
13
# with respect to perturbations in b. The `realpart` of the objective function
14
#
15
-# f(λ) = λ - 1,
+# f(λ(b)) = λ(b) - 1,
16
17
-# extracts the difference, whereas the `dualpart` evaluates to f'(λ).
+# extracts the difference, whereas the `dualpart` evaluates to ∂f/∂b (λ(b)).
18
19
20
n = 100
21
-λ = dual(0.0,0.0)
22
-b = dual(eps(), 1.0)
+λ = dual(0.0, 0.0)
+b = dual(0.0, 1.0)
23
v = zeros(Dual{Float64}, n)
24
v[1] = 1
25
v .+= 0.003.*randn(Float64, n)
0 commit comments