The PR code in Validation/PR/make_pr.py was built with K-shortest-paths-like algorithms in mind. It does not currently work unless the third column of an edges file is titled KSP index.
I see three possible fixes:
index DataFrames by column number, rather than column name
add in new column names to the code as they are needed
only make use of the third column if there is a a column titled KSP index.
I'm inclined to go down route 3, since it makes it easiest on the user of the software.
The PR code in
Validation/PR/make_pr.py
was built with K-shortest-paths-like algorithms in mind. It does not currently work unless the third column of an edges file is titledKSP index
.I see three possible fixes:
KSP index
.I'm inclined to go down route 3, since it makes it easiest on the user of the software.