v2.3 patch notes (#76)

* Docs: Adds v2.3 changelog

* Cleanup: Minor imports and style

docs/source/changelog.rst

@@ -1,7 +1,34 @@
 Changelog
 =========
 
-2.2.0 / 2018-MM-DD
+2.3.0 / 2018-12-01
+------------------
+
+This release primarily focuses on expanding the suite of available path
+technologies to provide better optimization characistics for 4-20 tensors while
+decreasing the time to find paths for 50-200+ tensors. See `Path Overview <path_finding.html#performance-comparison>`_ for more information.
+
+New Features
+++++++++++++
+- (:pr:`60`) A new ``greedy`` implementation has been added which is up to two orders of magnitude faster for 200 tensors.
+- (:pr:`73`) Adds a new ``branch`` path that uses ``greedy`` ideas to prune the ``optimal`` exploration space to provide a better path than ``greedy`` at sub ``optimal`` cost.
+- (:pr:`73`) Adds a new ``auto`` keyword to the :func:`opt_einsum.contract` ``path`` option. This keyword automatically chooses the best path technology that takes under 1ms to execute.
+
+Enhancements
+++++++++++++
+- (:pr:`61`) The :func:`opt_einsum.contract` ``path`` keyword has been changed to ``optimize`` to more closely match NumPy. ``path`` will be deprecated in the future.
+- (:pr:`61`) The :func:`opt_einsum.contract_path` now returns a :func:`opt_einsum.contract.PathInfo` object that can be queried for the scaling, flops, and intermediates of the path. The print representation of this object is identical to before. 
+- (:pr:`61`) The default ``memory_limit`` is now unlimited by default based on community feedback.
+- (:pr:`66`) The Torch backend will now use ``tensordot`` when using a version of Torch which includes this functionality.
+- (:pr:`68`) Indices can now be any hashable object when provided in the `"Interleaved Input" <input_format.html#interleaved-input>`_ syntax.
+- (:pr:`74`) Allows the default `transpose` operation to be overridden to take advantage of more advanced tensor transpose libraries.
+- (:pr:`73`) The ``optimal`` path is now significantly faster.
+
+Bug fixes
++++++++++
+- (:pr:`72`) Fixes the `"Interleaved Input" <input_format.html#interleaved-input>`_ syntax and adds documentation.
+
+2.2.0 / 2018-07-29
 ------------------
 
 New Features

opt_einsum/backends/__init__.py

@@ -2,10 +2,9 @@
 Compute backends for opt_einsum.
 """
 
-from .dispatch import (get_func, has_einsum, build_expression, evaluate_constants, has_backend)
-
 # Backends
 from .cupy import to_cupy
+from .dispatch import (get_func, has_einsum, build_expression, evaluate_constants, has_backend)
 from .tensorflow import to_tensorflow
 from .theano import to_theano
 from .torch import to_torch

opt_einsum/helpers.py

@@ -3,6 +3,7 @@
 """
 
 import numpy as np
+
 from .parser import get_symbol
 
 __all__ = ["build_views", "compute_size_by_dict", "find_contraction", "flop_count"]
@@ -18,9 +19,9 @@ def build_views(string, dimension_dict=None):
 
     Parameters
     ----------
-    tensor_list : list of str
+    string : list of str
         List of tensor strings to build
-    dimension_dictionary : dictionary
+    dimension_dict : dictionary
         Dictionary of index _sizes
 
     Returns

opt_einsum/tests/test_contract.py

@@ -96,7 +96,6 @@
     'aef,fbc,dca->bde',
 ]
 
-
 all_optimizers = ['optimal', 'branch-all', 'branch-2', 'branch-1', 'greedy']
 
 
@@ -181,8 +180,7 @@ def test_contract_expressions(string, optimize, use_blas, out_spec):
     shapes = [view.shape for view in views]
     expected = contract(string, *views, optimize=False, use_blas=False)
 
-    expr = contract_expression(
-        string, *shapes, optimize=optimize, use_blas=use_blas)
+    expr = contract_expression(string, *shapes, optimize=optimize, use_blas=use_blas)
 
     if out_spec and ("->" in string) and (string[-2:] != "->"):
         out, = helpers.build_views(string.split('->')[1])

opt_einsum/tests/test_input.py

@@ -2,8 +2,6 @@
 Tests the input parsing for opt_einsum. Duplicates the np.einsum input tests.
 """
 
-import sys
-
 import numpy as np
 import pytest