You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
From Wikipedia : https://en.wikipedia.org/wiki/MPS_(format) The optional BOUNDS section specifies lower and upper bounds on individual variables, if they are not given by rows in the matrix. .... Bound type PL is for a free positive from zero to plus infinity, but as this is the normal default, it is seldom used.
About 10% of models in MipLib2017 benchmark contain PL Bound variable lines and PuLP throws an exception when running var, prob = LpProblem.fromMPS(modelName)
The fix appears to be easy: in mps_lp.py, under readMPSSetBounds add two lines. There is no need to call set_both_bounds because the bounds are <0, infinity> which is the default
if bound == "FR":
set_both_bounds(None, None)
return
elif bound == "BV":
set_both_bounds(0, 1)
return
# add these two lines
elif bound == "PL":
return
The text was updated successfully, but these errors were encountered:
System:
MacBook (INTEL CPU), Ventura 13.2, Python 3.8
PuLP 2.7.0
From Wikipedia :
https://en.wikipedia.org/wiki/MPS_(format)
The optional BOUNDS section specifies lower and upper bounds on individual variables, if they are not given by rows in the matrix. .... Bound type PL is for a free positive from zero to plus infinity, but as this is the normal default, it is seldom used.
About 10% of models in MipLib2017 benchmark contain
PL Bound variable
lines and PuLP throws an exception when runningvar, prob = LpProblem.fromMPS(modelName)
The fix appears to be easy: in mps_lp.py, under
readMPSSetBounds
add two lines. There is no need to callset_both_bounds
because the bounds are <0, infinity> which is the defaultThe text was updated successfully, but these errors were encountered: