exponential function is doubled as a function of infinite parameters

[julia-sand]

Hi I have a problem issue with using the exponential function within infiniteopt models.

Describe the bug
an exponential function with infinite parameters in the infinite model causes the exponential function to be doubled in the output.

To Reproduce

using InfiniteOpt

#Setup the model
#create a new model object
model = InfiniteModel();

#add any infinite parameter
@infinite_parameter(model, q in [-5, 5], num_supports = 11)

#try any function with an exponential
print(exp(-(q-1)^4))

output then has the exponential doubled

exp(-exp((q - 1)^4)

Desktop (please complete the following information):

• OS linux
• InfiniteOpt v0.5.9
• Julia Version: both in 1.11.3 and 1.10.4

[odow]

It looks like this is just a bug in the printing. The actual numeric values are correct.

julia> using InfiniteOpt, Ipopt

julia> model = InfiniteModel(Ipopt.Optimizer);

julia> @infinite_parameter(model, q in [0, 1]);

julia> optimize!(model)

julia> y = exp(-q^3)
exp(-exp(q^3)

julia> value(y)
10-element Vector{Float64}:
 1.0
 0.9986291982956845
 0.9890860570884186
 0.9636404443013933
 0.9159518371704141
 0.8424274240215702
 0.7435670792055757
 0.6246851322548268
 0.49542863511370566
 0.36787944117144233

julia> exp.(-value(q).^3)
10-element Vector{Float64}:
 1.0
 0.9986291982956845
 0.9890860570884186
 0.9636404443013933
 0.9159518371704141
 0.8424274240215702
 0.7435670792055757
 0.6246851322548268
 0.49542863511370566
 0.36787944117144233

julia> exp.(-exp.(value(q).^3))
10-element Vector{Float64}:
 0.36787944117144233
 0.36737480560798674
 0.36384243664781807
 0.354257420122658
 0.3356251183824255
 0.3051212999827777
 0.2605739509203343
 0.2017337606088849
 0.13286068497591894
 0.06598803584531254

[pulsipher]

Yep, this is just a printing bug in InfiniteOpt's legacy NL interface. The developmental version no longer has this problem. My schedule has been busy, but I hope to finish the next release soon.

[pulsipher]

I am closing this since v0.6 fixes this