Michael Hippke
Michael Hippke
I see two requirements here: - Vectorize to get rid of the loop (make the code prettier) - Vectorize to gain speed. I can add a loop "inside" wotan (in...
If you're working with XY+brightness pixel data, then wotan is not the right tool. It's build for time series data (time, flux, errors). Try [this function](https://docs.astropy.org/en/stable/api/astropy.stats.biweight_location.html) instead. Regarding your points:...
You could shift your light curves so that they center around one?
- Added own lowess implementation instead of statsmodels dependency - faster (based on numba) - cleaner code (38 sloc instead of 678 sloc in statsmodels) - convergence to defined level...
lowess released with 1.0.7. estimators Hampel, Huber, Ramsay still open
I'm no expert of GPs. What do you recommend to improve the situation?
Thank you for the quick reply! `SliceSampler` with `generate_direction=generate_cube_oriented_direction` works, but is very slow (~50x fewer evals/sec compared to ReactiveNestedSampler). My problem requires ~2e8 evals until convergence with the StepSampler....
What about a simpler approach that runs in linear time and is essentially instant for
Can you check the [example](https://github.com/hippke/wotan/blob/master/examples/splines.ipynb) which has a huber spline application? In the example it works fine.
Yes, if there is a good case of what needs to be developed?