Add change of default AD backend

src/CTBase.jl

@@ -236,7 +236,8 @@ export Description, add, getFullDescription
 export CTException, ParsingError, AmbiguousDescription, IncorrectMethod
 export IncorrectArgument, IncorrectOutput, NotImplemented, UnauthorizedCall
 
-# checking
+# AD
+export set_AD_backend
 
 # functions
 export Hamiltonian, HamiltonianVectorField, VectorField

src/default.jl

@@ -1,11 +1,19 @@
+#
+__default_AD_backend = AutoForwardDiff()
+
+function set_AD_backend(AD) 
+    global __default_AD_backend = AD
+    nothing
+end
+
 """
 $(TYPEDSIGNATURES)
 
 Used to set the default value of Automatic Differentiation backend.
 
 The default value is `AutoForwardDiff()`, that is the `ForwardDiff` package is used by default.
 """
-__auto() = AutoForwardDiff() # default AD backend
+__get_AD_backend() = __default_AD_backend # default AD backend
 
 """
 $(TYPEDSIGNATURES)

src/repl.jl

@@ -14,9 +14,9 @@ end
 end
 
 #
-global ct_repl_is_set::Bool = false
-global ct_repl_data::CTRepl 
-global ct_repl_ct_repl_history::HistoryRepl
+ct_repl_is_set::Bool = false
+ct_repl_data::CTRepl = CTRepl()
+ct_repl_history::HistoryRepl = HistoryRepl(0, Vector{ModelRepl}())
 
 """
 $(TYPEDSIGNATURES)
@@ -47,13 +47,14 @@ Create a ct REPL.
 """
 function ct_repl(; debug=false, demo=false, verbose=false)
 
+    global ct_repl_is_set
+    global ct_repl_data
+    global ct_repl_history
+
     if !ct_repl_is_set
-#
-        global ct_repl_is_set = true
-
-        # init: ct_repl_data, ct_repl_history
-        global ct_repl_data = CTRepl()
-        global ct_repl_history = HistoryRepl(0, Vector{ModelRepl}())
+
+        #
+        ct_repl_is_set = true
 
         #
         ct_repl_data.debug = debug

src/utils.jl

@@ -73,7 +73,7 @@ $(TYPEDSIGNATURES)
 
 Return the gradient of `f` at `x`.
 """
-function ctgradient(f::Function, x::ctNumber; backend=__auto())
+function ctgradient(f::Function, x::ctNumber; backend=__get_AD_backend())
     extras = prepare_derivative(f, backend, x)
     return derivative(f, backend, x, extras)
 end
@@ -83,7 +83,7 @@ $(TYPEDSIGNATURES)
 
 Return the gradient of `f` at `x`.
 """
-function ctgradient(f::Function, x; backend=__auto())
+function ctgradient(f::Function, x; backend=__get_AD_backend())
     extras = prepare_gradient(f, backend, x)
     return gradient(f, backend, x, extras)
 end
@@ -100,7 +100,7 @@ $(TYPEDSIGNATURES)
 
 Return the Jacobian of `f` at `x`.
 """
-function ctjacobian(f::Function, x::ctNumber; backend=__auto())
+function ctjacobian(f::Function, x::ctNumber; backend=__get_AD_backend())
     f_number_to_number = only ∘ f ∘ only
     extras = prepare_derivative(f_number_to_number, backend, x)
     der = derivative(f_number_to_number, backend, x, extras)
@@ -112,7 +112,7 @@ $(TYPEDSIGNATURES)
 
 Return the Jacobian of `f` at `x`.
 """
-function ctjacobian(f::Function, x; backend=__auto())
+function ctjacobian(f::Function, x; backend=__get_AD_backend())
     extras = prepare_jacobian(f, backend, x)
     return jacobian(f, backend, x, extras)
 end

test/test_description.jl

@@ -7,9 +7,6 @@ descriptions = add(descriptions, (:b,))
 @test descriptions[1] == (:a,)
 @test descriptions[2] == (:b,)
 
-# print a tuple of descriptions
-@test display(descriptions) isa Nothing
-
 # get the complete description of the chosen method
 algorithmes = ()
 algorithmes = add(algorithmes, (:descent, :bfgs, :bissection))