ENH: implement any and all functions #175
Conversation
|
Great effort. I've left a few comments. Overall, this looks great, just needs some tweaks, namely in the tests and docs.
You will need to generate the docs once in order to produce |
| @@ -811,6 +811,127 @@ def max(self, axis=None, keepdims=False, out=None): | |||
| """ | |||
| return np.maximum.reduce(self, out=out, axis=axis, keepdims=keepdims) | |||
|
|
|||
| def any(self, axis=None, keepdims=False, out=None): | |||
| """ | |||
| Minimize along the given axes. Uses all axes by default. | |||
hameerabbasi
Jul 26, 2018
Collaborator
Needs to be amended.
Needs to be amended.
| -------- | ||
| :obj:`numpy.min` : Equivalent numpy function. | ||
| scipy.sparse.coo_matrix.min : Equivalent Scipy function. | ||
| :obj:`nanmin` : Function with ``NaN`` skipping. |
hameerabbasi
Jul 26, 2018
Collaborator
The See also section needs amending.
The See also section needs amending.
| >>> x = np.add.outer(np.arange(5), np.arange(5)) | ||
| >>> x # doctest: +NORMALIZE_WHITESPACE | ||
| array([[0, 1, 2, 3, 4], |
hameerabbasi
Jul 26, 2018
Collaborator
Should ideally be a boolean array that's sometimes false and sometimes true.
Should ideally be a boolean array that's sometimes false and sometimes true.
| See Also | ||
| -------- | ||
| :obj:`numpy.all` : Equivalent numpy function. | ||
| scipy.sparse.coo_matrix.all : Equivalent Scipy function. |
hameerabbasi
Jul 26, 2018
Collaborator
This doesn't exist, I'd cite any and all from each other.
This doesn't exist, I'd cite any and all from each other.
| >>> x = np.all.outer(np.arange(5), np.arange(5)) | ||
| >>> x # doctest: +NORMALIZE_WHITESPACE | ||
| array([[0, 1, 2, 3, 4], |
hameerabbasi
Jul 26, 2018
Collaborator
Same here. Ideally use this array for both to show the output, it's like a truth table:
x = np.array([
[False, False],
[False, True],
[True, False],
[False, False]
])
(It could be formatted better)
Same here. Ideally use this array for both to show the output, it's like a truth table:
x = np.array([
[False, False],
[False, True],
[True, False],
[False, False]
])(It could be formatted better)
| @@ -16,6 +16,8 @@ | |||
| ('sum', {'dtype': np.float16}, {'atol': 1e-2}), | |||
| ('prod', {}, {}), | |||
| ('min', {}, {}), | |||
| ('all', {}, {}), | |||
hameerabbasi
Jul 26, 2018
Collaborator
We'll need separate tests, input should be np.bool_ arrays. Take a look at how we've produced random arrays for binary ops and use that.
We'll need separate tests, input should be np.bool_ arrays. Take a look at how we've produced random arrays for binary ops and use that.
|
Oh. Apparently you're using an old NumPy version... The docstrings need to be produced with newer versions to pass the tests, unfortunately. (1.14.0 or newer I think). |
|
@stsievert Are you still interested in working on this? If not, is it okay if I pushed to your branch? You'd still be listed as the author of the commit. |
That's fine by me. |
Codecov Report
@@ Coverage Diff @@
## master #175 +/- ##
==========================================
+ Coverage 97.14% 97.22% +0.07%
==========================================
Files 11 11
Lines 1259 1295 +36
==========================================
+ Hits 1223 1259 +36
Misses 36 36
Continue to review full report at Codecov.
|
|
It seems like NumPy 1.15 changed the docstrings for bool arrays, omitting the dtype. For the tests to pass both locally and in Travis, I needed to skip these doctests. |
This PR closes #173 and adds
anyandallmethods toCOO.