search

dendrograms / astrodendro

Generate a dendrogram from a dataset
https://dendrograms.readthedocs.io/
Other
37 stars 38 forks source link

Add area to available statistics #70

Closed astrofrog closed 10 years ago

astrofrog commented 11 years ago

Should be straightforward - need to remember to get the units right if spatial_scale is available.

ChrisBeaumont commented 11 years ago

Good idea. There are a few definitions for area in common use -- one is just the area of the mask (or projected mask in PPV). The other uses sig_maj and sig_min, to give an intensity-weighted spatial dispersion. It's probably best to just pick one for now.

astrofrog commented 11 years ago

I was thinking more the actual area of the mask, which is the 'real' area of the structure. The area of the ellipse is easy enough to derive. Though we could have area_exact and area_ellipse or something like this?

astrofrog commented 10 years ago

@ChrisBeaumont - what do you think of the attached code? For area_exact for the PPV statistic I was thinking of finding the projected pixel area if the structure is collapsed in the velocity direction. Does this make sense? If so I can add it.

ChrisBeaumont commented 10 years ago

+1. Projecting the PPV cube to compute area_exact sounds good -- Rahul Shetty (and others) have used this definition of area in papers.

ChrisBeaumont commented 10 years ago

Looks good, except for the note about the imprecise docstring

astrofrog commented 10 years ago

I clarified the docstring. I didn't include the equation because it won't be visible in the docs since this is a property (only the first sentence of text is shown).

ChrisBeaumont commented 10 years ago

Looks good (is the full docstring at least displayed by IPython?)

astrofrog commented 10 years ago

@ChrisBeaumont - yes, it should be actually - though in that case I would vote that we can include the equation as 'human-readable' rather than LaTeX since it will never be rendered in LaTeX.

ChrisBeaumont commented 10 years ago

Looks good to me