The constructor for triangular distributions, Tri(a, b, c) is calling scipy.stats.triang as follows:
return uv(rv=ss.triang(c, loc=a, scale=b-a), tag=tag)
where a and b are respectively the min and max of the distribution and c is the peak. However, even if this is not clear from the SciPy documentation, the first argument to scipy.stats.triang should be between 0 and 1. That line should read something like
return uv(rv=ss.triang((c-a)/float(b-a), loc=a, scale=b-a), tag=tag)
The constructor for triangular distributions,
Tri(a, b, c)is callingscipy.stats.triangas follows:return uv(rv=ss.triang(c, loc=a, scale=b-a), tag=tag)where
aandbare respectively the min and max of the distribution andcis the peak. However, even if this is not clear from the SciPy documentation, the first argument toscipy.stats.triangshould be between 0 and 1. That line should read something likereturn uv(rv=ss.triang((c-a)/float(b-a), loc=a, scale=b-a), tag=tag)