Atreyee Sinha

@adonath So you are suggesting to have `FluxPoints.guess_sed_type()` ?

I added one item to the documentation list, ie, clarification of pulsed analysis. The current pulsar analysis notebook simply mentions that the phasing has to be done outside gammapy, but...

Probably the centre of the map is not a good default choice? If an observation is passed, the pointing position (or a fixed offset from it) should be used. Else,...

@bkhelifi for the case what you term "RoI analysis", I think we **always** encourage users to work with cutouts, ie, inside the data reduction loop, use `empty = stacked.cutout(obs.pointing_radec, width=2...

Currently `MapEvaluator` is not visible in the public API. Maybe making it visible will make the peek more useful.

Reviving this issue again - can we decide on the content of the meta table for 1.0 ? The current `meta_table` is not very useful (even though it has a...

What would be a good way to do it? Add an information in `meta` (but we decided not to use the meta_table internally in gammapy anywhere). While serialising through `Datasets`...

Indeed it will not. Sorry for the noise. I missed it somehow. But why do we add the LightCurve prefix for this specific model? It makes it a bit hard...

Thank you @registerrier ! The final request is apparently indeed for simulating time dependent spectral variability. I suppose this might be more general than just for TemplateModels. Maybe we can...

This was discussed at length today during the co-working week, and the plan forward is: - [x] Implement numerical implementation on `TemporalModel` @adonath #3925 - [ ] Adapt temporal models...