Closed brynpickering closed 7 months ago
All modified and coverable lines are covered by tests :white_check_mark:
Comparison is base (
4e77873
) 95.77% compared to head (69bccd9
) 95.83%.
Fantastic addition! Adding a list of the constraints where a parameter is used is a really good idea. However, it seems like some math is broken in the new parameters section. Here's an example:
OK great, that was a missing $
in the string. If you find any others, do flag them!
Chosen a subset of parameters listed in the inter-cluster storage docs. Also updated links in docs to prefer linking to the math docs parameter list rather than the schema. Annoyingly, because the math docs are auto-generated it isn't possible to link directly to the Parameters
anchor in the math docs.
Fixes issue mentioned in #559 w.r.t. information provided in the pre-defined math docs.
Summary of changes in this pull request:
where
string of other math components, not only in the math expression.defaults
which are used to initialise empty array elements. This helps handle situations where a sparsely defined model would lead to NaN values that shouldn't make their way into the LP file. The values are set such that the variable/expression does not impact the optimisation problem if taking its "default" value.where
strings to decide whether some variables/expressions/constraints need to be re-built. This should be expected functionality (i.e., if updating a parameter turns a NaN input value into a non-NaN value, then thewhere
string should be re-evaluated).TODO:
Reviewer checklist: