Sundy doc changes #5360
Sundy doc changes #5360
Conversation
change '{ndarray, float}' -> 'ndarray or float' as {} are for
when the value is an enumeration
- {ndarray, bool} -> ndarray or bool
- {int, array of ints} -> int or array of ints
- {str, function} -> str or function
- {sequnce, int} -> sequence or int
- {array_like, poly1d} -> array_like or poly1d
- {None, list of `m` scalars, scalar} -> list of `m` scalars or scalar
| @@ -1120,7 +1120,7 @@ def tensordot(a, b, axes=2): | |||
| ---------- | |||
| a, b : array_like, len(shape) >= 1 | |||
| Tensors to "dot". | |||
| axes : variable type | |||
| axes : integer or (2,) array_like | |||
There was a problem hiding this comment.
nitpick, should be int not integer
|
This looks good to me. |
|
Should I be worried about the two failed tests? |
|
no, travis just updated their numpy and I forgot to bump the version on the branch |
|
I assume you also want the other doc pr retargetted too? On Wed, Dec 10, 2014, 13:29 Julian Taylor notifications@github.com wrote:
|
|
mh I think you can't change target in github, we might as well then merge it manually. |
|
unfortunately it conflicts with other doc changes on master |
|
I can re-base to current master (and combine with #5359) tonight. Out of curiosity why do you want this on master rather than maint? For mpl I would have considered these changes 'bug fixes'. |
|
numpy uses the development on master -> backport to maintenance model |
|
Fair enough. |
| @@ -148,7 +148,7 @@ def pmt(rate, nper, pv, fv=0, when='end'): | |||
| Number of compounding periods | |||
| pv : array_like | |||
| Present value | |||
| fv : array_like (optional) | |||
| fv : array_like, (optional) | |||
There was a problem hiding this comment.
Parenthesis should be removed for optional to match the other docstrings
All documentation changes, mostly removing the abuse of {} for multiple return types.