fixes for 0.5 function changes

test/brent.jl

@@ -1,6 +1,6 @@
-f(x) = 2x^2+3x+1
+f_b(x) = 2x^2+3x+1
 
-results = optimize(f, -2.0, 1.0, method = :brent)
+results = optimize(f_b, -2.0, 1.0, method = :brent)
 
 @assert results.converged
 @assert abs(results.minimum+0.75) < 1e-7

test/callbacks.jl

@@ -1,12 +1,4 @@
 
-function cb(tr::OptimizationTrace)
-    @test tr.states[end].iteration % 3 == 0
-end
-
-function cb(os::OptimizationState)
-    @test os.iteration % 3 == 0
-end
-
 for method in (:nelder_mead,
                :simulated_annealing)
     ot_run = false

test/gradient_descent.jl

@@ -1,14 +1,14 @@
-function f_gd(x)
+function f_gd_1(x)
   (x[1] - 5.0)^2
 end
 
-function g_gd(x, storage)
+function g_gd_1(x, storage)
   storage[1] = 2.0 * (x[1] - 5.0)
 end
 
 initial_x = [0.0]
 
-d = DifferentiableFunction(f_gd, g_gd)
+d = DifferentiableFunction(f_gd_1, g_gd_1)
 
 results = Optim.gradient_descent(d, initial_x)
 @assert isempty(results.trace.states)
@@ -17,16 +17,16 @@ results = Optim.gradient_descent(d, initial_x)
 
 eta = 0.9
 
-function f_gd(x)
+function f_gd_2(x)
   (1.0 / 2.0) * (x[1]^2 + eta * x[2]^2)
 end
 
-function g_gd(x, storage)
+function g_gd_2(x, storage)
   storage[1] = x[1]
   storage[2] = eta * x[2]
 end
 
-d = DifferentiableFunction(f_gd, g_gd)
+d = DifferentiableFunction(f_gd_2, g_gd_2)
 
 results = Optim.gradient_descent(d, [1.0, 1.0])
 @assert isempty(results.trace.states)

test/nelder_mead.jl

@@ -9,23 +9,23 @@ for (name, prob) in Optim.UnconstrainedProblems.examples
 	end
 end
 
-function f(x::Vector)
+function f_nm(x::Vector)
   (100.0 - x[1])^2 + x[2]^2
 end
 
-function rosenbrock(x::Vector)
+function rosenbrock_nm(x::Vector)
   (1.0 - x[1])^2 + 100.0 * (x[2] - x[1]^2)^2
 end
 
 initial_x = [0.0, 0.0]
 
-results = Optim.nelder_mead(f, initial_x)
+results = Optim.nelder_mead(f_nm, initial_x)
 
 @assert results.f_converged
 @assert norm(results.minimum - [100.0, 0.0]) < 0.01
 @assert length(results.trace.states) == 0
 
-results = Optim.nelder_mead(rosenbrock, initial_x)
+results = Optim.nelder_mead(rosenbrock_nm, initial_x)
 
 @assert results.f_converged
 @assert norm(results.minimum - [1.0, 1.0]) < 0.01

test/newton.jl

@@ -1,18 +1,18 @@
 using Optim
 
-function f(x::Vector)
+function f_1(x::Vector)
     (x[1] - 5.0)^4
 end
 
-function g!(x::Vector, storage::Vector)
+function g!_1(x::Vector, storage::Vector)
     storage[1] = 4.0 * (x[1] - 5.0)^3
 end
 
-function h!(x::Vector, storage::Matrix)
+function h!_1(x::Vector, storage::Matrix)
     storage[1, 1] = 12.0 * (x[1] - 5.0)^2
 end
 
-d = TwiceDifferentiableFunction(f, g!, h!)
+d = TwiceDifferentiableFunction(f_1, g!_1, h!_1)
 
 results = Optim.newton(d, [0.0])
 @assert length(results.trace.states) == 0
@@ -21,23 +21,23 @@ results = Optim.newton(d, [0.0])
 
 eta = 0.9
 
-function f(x::Vector)
+function f_2(x::Vector)
   (1.0 / 2.0) * (x[1]^2 + eta * x[2]^2)
 end
 
-function g!(x::Vector, storage::Vector)
+function g!_2(x::Vector, storage::Vector)
   storage[1] = x[1]
   storage[2] = eta * x[2]
 end
 
-function h!(x::Vector, storage::Matrix)
+function h!_2(x::Vector, storage::Matrix)
   storage[1, 1] = 1.0
   storage[1, 2] = 0.0
   storage[2, 1] = 0.0
   storage[2, 2] = eta
 end
 
-d = TwiceDifferentiableFunction(f, g!, h!)
+d = TwiceDifferentiableFunction(f_2, g!_2, h!_2)
 results = Optim.newton(d, [127.0, 921.0])
 @assert length(results.trace.states) == 0
 @assert results.gr_converged

test/simulated_annealing.jl

@@ -1,15 +1,15 @@
 srand(1)
 
-function f(x::Vector)
+function f_s(x::Vector)
     (x[1] - 5.0)^4
 end
 
-results = Optim.simulated_annealing(f, [0.0])
+results = Optim.simulated_annealing(f_s, [0.0])
 @assert norm(results.minimum - [5.0]) < 0.1
 
-function rosenbrock(x::Vector)
+function rosenbrock_s(x::Vector)
     (1.0 - x[1])^2 + 100.0 * (x[2] - x[1]^2)^2
 end
 
-results = Optim.simulated_annealing(rosenbrock, [0.0, 0.0])
+results = Optim.simulated_annealing(rosenbrock_s, [0.0, 0.0])
 @assert norm(results.minimum - [1.0, 1.0]) < 0.1