Closed tal-nc closed 4 years ago
added another small fix for odd numbers. noticed there were differences between running with other connection only, vector alone vs the matrix alone. looked into it a bit and i saw that the matrix generated too many connections. it lead me there. it should be (and used to be) divided by 2. it got broke recently.
i am testing this by running on a custom population and comparing the matrix total to the expected one
It loos OK. How did you notice the bug and how did you check that this fixed it?