deletions | additions       diff --git a/section_Requirements_We_want_the__.tex b/section_Requirements_We_want_the__.tex  index ca93929..b5921c8 100644  --- a/section_Requirements_We_want_the__.tex  +++ b/section_Requirements_We_want_the__.tex 
...
I choose a square exponential kernel ({\tt ExpSquaredKernel} in {\tt george}) $$k(d)=θ_1^2 \exp(\frac{−d^2}{θ_2^2})$$.  In order to choose the hyperparameters $θ_1^2$ and $θ_2^2$ I minimize the objective function: $\chi^2(m) ~*~ \sum \sqrt{\sum_t  { \left| \sqrt{\frac{d^2 \frac{d^2  m}{dt^2} }\right| }$, where $m$ is the magnitude, and the fit is done in log time. This works remarkably well in general, with some ``pathological'' exceptions.