Open jsdillon opened 6 years ago
Big plus one from me. I had a really hard time figuring this part out and I never ended up with a useful legend. Even when I set average_times=True
I got 6 identical legend entries, but only 1 line showed, so I guess they were all on top of each other.
Also, the docs are kind of unhelpful, I never figured out that it wanted [((1,2),(3,4))]
, I just used this from the docs: [list(pspec.blpair_array),]
which was limiting. And the docs are slightly wrong in that they show using keywords for spw, pol rather than positional arguments.
https://github.com/HERA-Team/hera_pspec/blob/5dd299f4bc0f951e487f2f194ebf291687e62318/hera_pspec/plot.py#L20
First, "List of baseline-pair tuples" sounds like something like
[(1,2),(3,4)]
, when what you really mean is[((1,2),(3,4))]
. Trying the former gives the rather unhelpfulTypeError: 'int' object has no attribute '__getitem__'
.Second it's not clear that by spectral window you mean spectral window index, especially since I created my uvp object by specifying channels in a spw_range. In general, it's a bit unintuitive to figure out what the spws are and what they mean by looking to the attributes of a uvp object.
Third the legend (at least in the design review warmup) is pretty useless: It seems like at the very least you might want to show the LST too if you're not going to be smart in trying to make the legend show what's actually changing between the plotted spectra.
More generally, it's odd to me that you must specify which blpairs, spw, and pol to plot (that the default is not all of them) but that you apparently cannot specify which LSTs to plot...only whether to plot them all or average.