Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | DOC: missingdata: Add example of a C-API function supporting NA masks | Mark Wiebe | 2011-08-27 | 4 | -55/+387 |
| | |||||
* | DOC: missingdata: Documenting C API for NA-masked arrays | Mark Wiebe | 2011-08-27 | 3 | -0/+240 |
| | |||||
* | DOC: nditer: Document NpyIter_IsFirstVisit function | Mark Wiebe | 2011-08-27 | 1 | -0/+22 |
| | |||||
* | ENH: missingdata: Create count_reduce_items function | Mark Wiebe | 2011-08-27 | 1 | -0/+1 |
| | | | | | | | This function either cheaply returns the product of the sizes of all the reduction axes, or counts the number of items which will be used in a reduction operation when skipna is True. Its purpose is to make it easy to do functions like np.mean and np.std. | ||||
* | ENH: missingdata: Rewrite PyArray_Concatenate to work with NA masks | Mark Wiebe | 2011-08-27 | 1 | -12/+14 |
| | | | | | It should also have less memory usage for heterogeneous inputs, because it no longer makes extra copies in that case. | ||||
* | DOC: missingdata: Add some NA mask info to the documentation | Mark Wiebe | 2011-08-27 | 2 | -50/+214 |
| | |||||
* | DOC: nditer: Add details about why 'readonly' is the operand default | Mark Wiebe | 2011-08-25 | 1 | -1/+5 |
| | |||||
* | DOC: nditer: Tweaks to the tutorial based on feedback from Chris | Mark Wiebe | 2011-08-25 | 1 | -25/+31 |
| | |||||
* | DOC: nditer: Add links to the nditer introductory doc to make it more ↵ | Mark Wiebe | 2011-08-25 | 2 | -1/+6 |
| | | | | discoverable | ||||
* | DOC: nditer: Finish up the nditer walkthrough with a Cython example | Mark Wiebe | 2011-08-25 | 2 | -37/+185 |
| | |||||
* | DOC: nditer: Add tutorial-style material covering more than one operand | Mark Wiebe | 2011-08-25 | 1 | -15/+243 |
| | |||||
* | DOC: nditer: Write tutorial-style introduction covering single-array iteration | Mark Wiebe | 2011-08-25 | 1 | -0/+329 |
| | |||||
* | DOCS: New ufunc creation docs | Chris Jordan-Squire | 2011-08-22 | 1 | -2/+2 |
| | |||||
* | Merge branch 'deprecate_array_field_access' | Mark Wiebe | 2011-07-26 | 1 | -1/+26 |
|\ | |||||
| * | DOC: core: Document the PyArray_SetBaseObject function | Mark Wiebe | 2011-07-22 | 1 | -1/+26 |
| | | |||||
* | | DOC: datetime: Update the docs to reflect busday_count change | Mark Wiebe | 2011-07-19 | 1 | -2/+1 |
| | | |||||
* | | DOC: datetime: hyperlink busday_count and is_busday functions in datetime doc | Mark Wiebe | 2011-07-19 | 1 | -2/+2 |
| | | |||||
* | | DOC: datetime: Add additional tutorial information. | Steve R. Hastings | 2011-07-19 | 1 | -27/+95 |
|/ | |||||
* | DOC: fix build issues (with latex or newest sphinx) | Pauli Virtanen | 2011-07-15 | 1 | -11/+7 |
| | |||||
* | DOC: core: Document the mask-based nditer flags and new inline mask functions | Mark Wiebe | 2011-07-11 | 3 | -10/+249 |
| | |||||
* | ENH: core: Add np.copyto, PyArray_MaskedMoveInto, PyArray_MaskedCopyInto | Mark Wiebe | 2011-07-08 | 2 | -0/+15 |
| | | | | | | | | These functions expose masked copying routines, with and without handling of overlapping data. Also deprecated the np.putmask and PyArray_PutMask functions, because np.copyto supercedes their functionality. This will need to be discussed on the list during the pull request review. | ||||
* | DOC: c-api: Update to reflect array flag namespace change | Mark Wiebe | 2011-07-06 | 3 | -144/+149 |
| | |||||
* | DOC: Document the ufunc 'where=' parameter and the NpyAuxData C API mechanism | Mark Wiebe | 2011-07-06 | 2 | -0/+99 |
| | |||||
* | DOC: datetime: Split the date/time units into two separate tables | Mark Wiebe | 2011-07-01 | 1 | -7/+24 |
| | |||||
* | DOC: datetime: Updates based on Chuck's feedback | Mark Wiebe | 2011-07-01 | 1 | -6/+16 |
| | |||||
* | DOC: datetime: Start a draft of introductory datetime documentation | Mark Wiebe | 2011-07-01 | 2 | -0/+262 |
| | |||||
* | DOC: list routines for dealing with raw binary files | Pauli Virtanen | 2011-06-25 | 1 | -0/+8 |
| | |||||
* | DOC: dtype-struct: Document the changes to struct dtypes | Mark Wiebe | 2011-06-22 | 1 | -38/+56 |
| | |||||
* | ENH: Add the polynomial module to the documentation. | Charles Harris | 2011-06-20 | 4 | -3/+33 |
| | | | | | | | | | | | Also: 1. Note that the polynomial package is preferred over poly1d. 2. Remove deprecation of mapparms in __init__.py as it interferes with the documention of the method of the same name. This is probably safe as it is unlikely to be used yet. 3. Make some improvements to the documentation in polytemplate. | ||||
* | STY: datetime-feedback: Rename np.busdaydef -> np.busdaycalendar | Mark Wiebe | 2011-06-15 | 1 | -1/+1 |
| | | | | | Also rename the busdaydef parameters to busdaycal parameters. This change was motivated by Chuck's code review feedback. | ||||
* | DOC: datetime-bday: Document the datetime business day functions | Mark Wiebe | 2011-06-14 | 2 | -0/+21 |
| | |||||
* | DOC: Document the new keyword arguments for ufuncs | Mark Wiebe | 2011-04-22 | 1 | -0/+39 |
| | |||||
* | DOC: Improve the documentation about type promotion | Mark Wiebe | 2011-04-22 | 2 | -7/+49 |
| | |||||
* | DOC: The nditer exposure wasn't linked correctly | Mark Wiebe | 2011-03-22 | 2 | -1/+1 |
| | |||||
* | API: Rename 'niter' to 'nop' in the nditer | Mark Wiebe | 2011-03-17 | 1 | -24/+24 |
| | | | | | | This name was chosen partially based on the previous multi-iter, which stored an array of iterators, so 'niter' made sense. In the new nditer, it doesn't, and 'nop' for number of operands seems better. | ||||
* | DOC: Fill in more of the nditer docs | Mark Wiebe | 2011-03-16 | 1 | -0/+1 |
| | |||||
* | ENH: Add 'subok' parameter to PyArray_NewLikeArray, np.empty_like, ↵ | Mark Wiebe | 2011-03-15 | 1 | -1/+5 |
| | | | | | | | np.zeros_like, and np.ones_like This way, the sub-type can be avoided if necessary. This helps mitigate, but doesn't fix, ticket #1753, by allowing "b = np.empty_like(a, subok=False)". | ||||
* | API: Rename 'coords' to 'multi-index' in ravel_coords and iterator API | Mark Wiebe | 2011-03-14 | 2 | -35/+35 |
| | |||||
* | DOC: BLD: fix some doc build warnings. | Ralf Gommers | 2011-03-14 | 2 | -2/+1 |
| | | | | | Word of warning: do not use Sphinx 1.0.7 (latest release), its autodoc extension is broken. 1.0.4 works fine. | ||||
* | STY: Rename NPY_ITER_NO_INNER_ITERATION to NPY_ITER_EXTERNAL_LOOP | Mark Wiebe | 2011-03-14 | 1 | -22/+23 |
| | | | | | It's a little bit shorter, and more intuitively expresses what the flag does. | ||||
* | DOC: Document constructor API change, and fill in some more missing ↵ | Mark Wiebe | 2011-03-13 | 3 | -53/+115 |
| | | | | documentation | ||||
* | BUG: For compatibility with 1.5, revert to permitting limited broadcasting ↵ | Mark Wiebe | 2011-03-11 | 1 | -1/+1 |
| | | | | | | | | | of the assignment output This change got Travis's -10 veto for 1.6. An unfortunate consequence of reverting this is that some of the broadcasting error messages get worse, but they're still no worse than in 1.5. | ||||
* | DOC: Replace 'deprecated' with 'superceded' in a few places, fix a typo. | Mark Wiebe | 2011-03-10 | 2 | -16/+17 |
| | |||||
* | DOC: Add some missing documentation, hyper-link the iterator documentation | Mark Wiebe | 2011-03-10 | 2 | -164/+348 |
| | |||||
* | API: Rename the iterator function pointer types to be more consistent with ↵ | Mark Wiebe | 2011-03-10 | 1 | -8/+8 |
| | | | | | | | NumPy convention 'NpyIter_IterNext_Fn' -> 'NpyIter_IterNextFunc *' 'NpyIter_GetCoords_Fn' -> 'NpyIter_GetCoordsFunc *' | ||||
* | API: Change iterator API parameters ndim and niter from npy_intp to int | Mark Wiebe | 2011-03-10 | 1 | -18/+18 |
| | | | | | These parameters are never large, so it's better to use a straight int instead of npy_intp, consistent with ndim in PyArrayObject as well. | ||||
* | DOC: Copy iterator API documentation from the NEP | Mark Wiebe | 2011-03-05 | 2 | -0/+1102 |
| | |||||
* | DOC: add some more wiki edits, of rst files in ref/user guide. | rgommers | 2011-03-02 | 3 | -10/+10 |
| | |||||
* | ENH: core: Allow user to pass in output array for dot() | Luis Pedro Coelho | 2011-02-12 | 1 | -1/+9 |
| | | | | | | | | | This avoids the memory allocation. It is strict in checking that the types are correct, but since it is intended as an optimisation, it should only be used when the user knows what they are doing. The out parameter is added both to the BLAS and non-BLAS versions of dot(). Tests are included. | ||||
* | ENH: index_tricks: Implement unravel_index and ravel_coords functions in C | Mark Wiebe | 2011-02-10 | 1 | -0/+1 |
| |