Add InfiniteInterpolate as an extension #382
Conversation
…linear interpolation + test case
…and handling for variables that don't rely on infinite parameters
…ng interpolated values
…linear interpolation + test case
…and handling for variables that don't rely on infinite parameters
…ng interpolated values
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## master #382 +/- ##
=======================================
Coverage 99.68% 99.69%
=======================================
Files 35 36 +1
Lines 6759 6786 +27
=======================================
+ Hits 6738 6765 +27
Misses 21 21 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
…interpolation method error
…linear interpolation + test case
…and handling for variables that don't rely on infinite parameters
…ng interpolated values
…linear interpolation + test case
…ng interpolated values
…interpolation method error
…ue func in hovercraft example + shorten doc string
…terpolate test case
|
Good work, before proceeding please address the failing test. I will work on a patch to fix the failing docs. |
|
@wenwen0231, as a workaround for this PR. I think we can clear the tests if we don't import interpolations at the start of |
…nterpolate test file
| @@ -20,12 +26,14 @@ JuMP = "1.18" | |||
| MutableArithmetics = "1" | |||
| Reexport = "0.2, 1" | |||
| julia = "^1.6" | |||
| Interpolations = "0.16.1" | |||
There was a problem hiding this comment.
Does it have to be this exact version, will other versions work? We should be permissive as possible.
| @@ -23,3 +24,4 @@ JuMP = "1.23" | |||
| Literate = "2.18" | |||
| Plots = "1" | |||
| SpecialFunctions = "2" | |||
| Interpolations = "0.16.1" | |||
There was a problem hiding this comment.
Please do 0.16 instead of 0.16.1, we shouldn't specify the patch release unless absolutely necessary, which I suspect is not the case here.
| @@ -61,15 +61,21 @@ end) | |||
| # Optimize the model: | |||
| optimize!(m) | |||
|
|
|||
| # Extract the results: | |||
| x_opt = value.(x); | |||
| # Extract the results. The InfiniteInterpolate extension can be used to get a smooth interpolated function for x, which is invoked when both the `Interpolations` and `InfiniteOpt` packages are imported: | |||
There was a problem hiding this comment.
I would expand a little more to say that you are choosing to do cubic splines here. Maybe also provide a link to the technical API documentation.
| 20.9999999956286 | ||
| ``` | ||
|
|
||
| !!! note |
There was a problem hiding this comment.
This should be a warning, not a note.
| @@ -137,7 +137,31 @@ parameter dependencies. | |||
| This also holds true for many other methods in `InfiniteOpt`. For example, | |||
| adding the dot also vectorizes `dual` and `set_binary`. | |||
|
|
|||
| We can also query the dual of a constraint via | |||
| We can also get a continuous representation of our variable as an interpolations object from the `Interpolations.jl` package. This is based on the `InfiniteInterpolate` extension, which is automatically loaded in when both `InfiniteOpt` and `Interpolations` are imported. Currently supported methods are `constant_interpolation`, `linear_interpolation` and `cubic_spline_interpolation`. | |||
There was a problem hiding this comment.
This should be probably be its own subsection with a header
| # Get the parameter supports | ||
| Vparams = [] | ||
| for pref in prefs | ||
| JuMP.check_belongs_to_model(pref, JuMP.owner_model(vref)) |
There was a problem hiding this comment.
This can never not be true. Please remove.
| if !(infOpt.UniformGrid in suppsLabel) && !(method isa irregularGridMethod) | ||
| throw(ArgumentError("Interpolation method $(method) does not support irregular grids for supports. Please specify a uniform grid or choose a different interpolation method.")) | ||
| end |
There was a problem hiding this comment.
I suspect this will need updating when the above piracy concerns are addressed.
| end | ||
|
|
||
| paramVals = infOpt._get_value(pref, infOpt._index_type(pref); kwargs...) | ||
| numSupps = infOpt.num_supports(pref) |
There was a problem hiding this comment.
Why do you call num_supports if you already have the values? Just use length instead.
ext/InfiniteInterpolate.jl
Outdated
| # Ensure Vparams is a tuple for interpolation | ||
| Vparams = Tuple(Vparams) | ||
|
|
||
| # Get the variable supports | ||
| Vsupps = infOpt._get_value(vref, infOpt._index_type(vref); kwargs...) | ||
|
|
||
| # Wrappers for interpolation method | ||
| interpolateFunc(interpMethod::Function, params::Tuple, supps::Vector) = interpMethod(params, supps) | ||
|
|
||
| # Wrapper for multi-parameter variables | ||
| interpolateFunc(interpMethod::Function, params::Tuple, supps::Matrix) = interpMethod(params, supps) | ||
|
|
||
| # Pass the parameter and variable values to interpolation function | ||
| varFunc = interpolateFunc(method, Vparams, Vsupps) | ||
| return varFunc | ||
| end | ||
| end |
| model = InfiniteModel(Optimizer) | ||
| tb = model.backend | ||
|
|
||
| # Define infinite parameters and variables |
There was a problem hiding this comment.
We should include testing what happens with orthogonal collocation points.
Summary
Add the module
InfiniteInterpolatewhich extends thevalue()function to return an interpolated function of the variable or parameter being queried, based on the interpolation method defined by the user.Closing issues
Fixes issue #82