Difference between paper and code

Hi,

I have been studying your paper, thanks for uploading your code with it !

I was wondering, when you apply SigmaConsensus, in the paper you use the formula (6) to compute weights, but in the code you only use the exponential part of this formula. Why is the factor \sigma^{-\rho} D(\theta, \sigma)^{\rho - 1} ignored in the computation of the weights ?

Best regards,
Clément Riu.