In a recent change to dot (https://github.com/bmorris3/dot/commit/3a39feb770aef653412e007ea9997212d0e71abe) I removed the mean from the fluxes and I remove the mean from the times. This causes an issue when computing the spot phi angle, which no longer has a reference time subtracted from the input times in the mean model. The fit to the light curve still looks good but the longitudes aren't constrained and the animation produced by movie doesn't match the light curve.
The fix is to implement a reference time in the model and set it to a sensible time within the bounds (min_time, max_time). PR incoming.
In a recent change to
dot(https://github.com/bmorris3/dot/commit/3a39feb770aef653412e007ea9997212d0e71abe) I removed the mean from the fluxes and I remove the mean from the times. This causes an issue when computing the spotphiangle, which no longer has a reference time subtracted from the input times in the mean model. The fit to the light curve still looks good but the longitudes aren't constrained and the animation produced bymoviedoesn't match the light curve.The fix is to implement a reference time in the model and set it to a sensible time within the bounds (
min_time,max_time). PR incoming.