search

rcfduarte / nmsat

Simulate and analyse complex microcircuits
GNU General Public License v2.0
6 stars 1 forks source link

Revised local variation #14

Closed zbarni closed 7 years ago

zbarni commented 7 years ago

The current formula in the code doesn't account for the singularity cases mentioned in the paper.. Should we change it to the expanded version (also in the paper)?

rcfduarte commented 7 years ago

you mean the lv_r? which singularity cases? I don't really remember this, as far as I remember, the lv_r was just regularized by the refractory time.. In any case, all the results I have so far show that the lv_r is always very similar to the lv, so it's not really a primary measure to use.. I would suggest keeping this issue open to handle later, but for now, it's not a main priority..

zbarni commented 7 years ago

As a result, the denominator of Equation 2, changes to . In order to avoid the singularity that may occur when is equal to or less than 2R, we performed a series expansion to the first order in R. The revised local variation LvR is thus defined as....

eq. 3 in the paper

rcfduarte commented 7 years ago

I think I tried to simplify the expression (eq. 3) when I implemented it.. I was just quickly checking the math and couldn't get to the same result, though.. I don't think the current implementation is wrong, since the results are always very consistent with the lv.. But in any case, it's probably better to double check.. But yeah, what I was aiming for in this implementation was equation 3 in the paper..