Fix up ActivationFunction definitions

Author: ChrisRackauckas

docs/src/examples/collocation.md

@@ -31,7 +31,7 @@ savefig("colloc.png")
 plot(tsteps,du')
 savefig("colloc_du.png")
 
-dudt2 = Lux.Chain(ActivationFunction(x -> x.^3),
+dudt2 = Lux.Chain(ActivationFunction(x -> x^3),
                   Lux.Dense(2, 50, tanh),
                   Lux.Dense(50, 2))
 
@@ -134,7 +134,7 @@ calculates the squared difference between `f(u,p,t)` and `u'` at each
 point, and find the parameters which minimize this difference:
 
 ```@example collocation
-dudt2 = Lux.Chain(ActivationFunction(x -> x.^3),
+dudt2 = Lux.Chain(ActivationFunction(x -> x^3),
                   Lux.Dense(2, 50, tanh),
                   Lux.Dense(50, 2))
 

docs/src/examples/multiple_shooting.md

@@ -41,7 +41,7 @@ ode_data = Array(solve(prob_trueode, Tsit5(), saveat = tsteps))
 
 
 # Define the Neural Network
-nn = Lux.Chain(ActivationFunction(x -> x.^3),
+nn = Lux.Chain(ActivationFunction(x -> x^3),
                   Lux.Dense(2, 16, tanh),
                   Lux.Dense(16, 2))
 p_init, st = Lux.setup(rng, nn)

docs/src/examples/neural_ode.md

@@ -28,7 +28,7 @@ end
 prob_trueode = ODEProblem(trueODEfunc, u0, tspan)
 ode_data = Array(solve(prob_trueode, Tsit5(), saveat = tsteps))
 
-dudt2 = Lux.Chain(ActivationFunction(x -> x.^3),
+dudt2 = Lux.Chain(ActivationFunction(x -> x^3),
                   Lux.Dense(2, 50, tanh),
                   Lux.Dense(50, 2))
 p, st = Lux.setup(rng, dudt2)
@@ -110,7 +110,7 @@ the layer. Here we're going to use `Lux.Chain`, which is a suitable neural netwo
 structure for NeuralODEs with separate handling of state variables:
 
 ```@example neuralode
-dudt2 = Lux.Chain(ActivationFunction(x -> x.^3),
+dudt2 = Lux.Chain(ActivationFunction(x -> x^3),
                   Lux.Dense(2, 50, tanh),
                   Lux.Dense(50, 2))
 p, st = Lux.setup(rng, dudt2)