ENH: Implement Panel pct_change #6909
Conversation
| @@ -1130,11 +1130,20 @@ def shift(self, lags, freq=None, axis='major'): | |||
| if freq: | |||
| return self.tshift(lags, freq, axis=axis) | |||
|
|
|||
| # TODO: why raise? axis=0 works, seems to be fine. need to chcek aligning | |||
There was a problem hiding this comment.
@jreback any idea why this isn't allowed? wp.shift(0) doesn't raise, and seems to produce the correct answer.
There was a problem hiding this comment.
their is a perf issue currently because of the recent change of panel.shift to generic.shift
see #6826
I think wp.shift(0) is technically ok (though not intuitive)
prob should have a panel.shift that enforces kwargs (and doesn't have positional args at all)
you can use some of the machinery in reindex to detect the correct signatures
There was a problem hiding this comment.
Ok. I'm going to hold off on touching Panel.shift() until #6826 is settled.
|
I added a vbench, but maybe it isn't worth it since really all that's happing is a panel.shift() and a panel.div(). Should I take it out? |
|
it's fine |
|
instead of specifying the items axis use info_axis for ndim > 2 maybe this should also be explict for ndim > 2 |
|
I changed the default axis to major to be more consistent with |
|
axis should default to stat_axis (which is 1 for a panel) |
Had to change the implementation in generic.pct_change to not pass `periods` as a keyword arg to generic. Once the signitures of DataFrame.shift() and Panel.shift() are aligned, this change can be reverted.
|
I finally understand what you meant about the axes (I've never really understood what |
|
gr8 - generic is good! makes maint easier u can simply put in the Panel4D notimementrd list (as this is prob not intuitive), look in panelnd for a list of functions that it disables |
|
Actually this will already raise NotImplemented since |
|
ok gr8 |
ENH: Implement Panel pct_change
Closes #6904
There's just a bit of extra index handling that needs to be done before moving on to
generic.pct_change(). I had to adjust that to use the.divand.subops instead of/and-to work with panels.I wasn't sure why axis wasn't included as an actual names keyword arg.
genericjust looks for it in **kwargs. I did the same in panel.A related issue was the
Panel.shift()has a different argument signature thangeneric.shift(). I can make those consistent and put in a deprecation warning in this issue or in a new one.