Merge commit 'v0.16.2-65-g054821d' into debian

* commit 'v0.16.2-65-g054821d':
  TST: 32-test fixes
  BUG: GH9907 generate_range when start and end have higher resolution than offset
  TST/CLN: remove assert_isinstance
  PERF: releasing the GIL, pandas-dev#8882
  PERF: vbenches for pandas-dev#8882, releasing the GIL

Author: yarikoptic

doc/source/whatsnew/v0.17.0.txt

@@ -13,6 +13,7 @@ users upgrade to this version.
 
 Highlights include:
 
+  - Release the Global Interpreter Lock (GIL) on some cython operations, see :ref:`here <whatsnew_0170.gil>`
 
 Check the :ref:`API Changes <whatsnew_0170.api>` and :ref:`deprecations <whatsnew_0170.deprecations>` before updating.
 
@@ -58,8 +59,32 @@ Deprecations
 Removal of prior version deprecations/changes
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
+.. _dask: https://dask.readthedocs.org/en/latest/
+
+.. _whatsnew_0170.gil:
+
+Releasing the GIL
+~~~~~~~~~~~~~~~~~
+
+We are releasing the global-interpreter-lock (GIL) on some cython operations.
+This will allow other threads to run simultaneously during computation, potentially allowing performance improvements
+from multi-threading. Notably ``groupby`` and some indexing operations are a benefit from this. (:issue:`8882`)
+
+For example the groupby expression in the following code will have the GIL released during the factorization step, e.g. ``df.groupby('key')``
+as well as the ``.sum()`` operation.
+
+.. code-block:: python
+
+   N = 1e6
+   df = DataFrame({'key' : np.random.randint(0,ngroups,size=N),
+                   'data' : np.random.randn(N) })
+   df.groupby('key')['data'].sum()
+
+Releasing of the GIL could benefit an application that uses threads for user interactions (e.g. ``QT``), or performaning multi-threaded computations. A nice example of a library that can handle these types of computation-in-parallel is the dask_ library.
+
 .. _whatsnew_0170.performance:
 
+
 Performance Improvements
 ~~~~~~~~~~~~~~~~~~~~~~~~
 - Added vbench benchmarks for alternative ExcelWriter engines and reading Excel files (:issue:`7171`)
@@ -90,12 +115,14 @@ Bug Fixes
 
 
 - Bug in ``Table.select_column`` where name is not preserved (:issue:`10392`)
+- Bug in ``offsets.generate_range`` where ``start`` and ``end`` have finer precision than ``offset`` (:issue:`9907`)
 
 
 - Bug in ``DataFrame.interpolate`` with ``axis=1`` and ``inplace=True`` (:issue:`10395`)
 
 
 
+
 - Bug in ``test_categorical`` on big-endian builds (:issue:`10425`)
 
 

pandas/core/common.py

@@ -839,7 +839,6 @@ def take_nd(arr, indexer, axis=0, out=None, fill_value=np.nan,
 
     func = _get_take_nd_function(arr.ndim, arr.dtype, out.dtype,
                                  axis=axis, mask_info=mask_info)
-
     indexer = _ensure_int64(indexer)
     func(arr, indexer, out, fill_value)