Conversation
| return Interval(a.lo * x, a.hi * x) | ||
| end | ||
| return Interval(a.hi * x, a.lo * x) | ||
| end |
There was a problem hiding this comment.
It's questionable that a very simple type like this and a few methods need a license attribution. Since both packages are MIT, we're in the clear, so I just moved into this file.
There was a problem hiding this comment.
I moved the files into the main one in MathOptConflict/Infeasibility/InfeasibleSubsystem, so I will not rush this part here for now
| for t in f.terms | ||
| out += _eval_variables(value_fn, t) | ||
| end | ||
| return out |
There was a problem hiding this comment.
Ditto here: why is this _eval_variales needed?
There was a problem hiding this comment.
because MOI assumes the output type is the same as the ScalarAffineFunction (Float64), but we need the output to be an interval. JuMP does is correctly: #19 (comment)
|
I was thining about code organization. The printing stuff is very verbose so I was considering splitting the analysis files in two:
|
|
replaced by #40 |
2 participants
I should probably split this up, but let me comment in-line