Merge 49f1e59 into 23314cb

CHANGELOG.rst

@@ -3,6 +3,10 @@ This file tracks important changes in PRs
 geoopt v0.1.0 (unreleased)
 ==========================
 
+Breaking Changes
+----------------
+* better public api, refactored developer api a lot (#40). See the corresponding PR for more details
+
 New Features
 ------------
 * Added ``Sphere`` manifold (#25)
@@ -11,7 +15,7 @@ New Features
 Maintenance
 -----------
 * Add gitter chat (#31)
-* Maintain torch>=1.0.0 only
+* Maintain torch>=1.0.0 only (#39)
 
 Deprecations
 ------------

README.rst

@@ -16,14 +16,14 @@ There are two ways to install geoopt:
 
 1. GitHub (preferred so far) due to active development
 
-.. code-block::
+.. code-block:: bash
 
     pip install git+https://github.com/ferrine/geoopt.git
 
 
 2. pypi (this might be significantly behind master branch)
 
-.. code-block::
+.. code-block:: bash
 
     pip install geoopt
 
@@ -54,11 +54,11 @@ points on a certain manifold
 -  ``.inner(u, v=None)`` – inner product at this point for two
    **tangent** vectors at this point. The passed vectors are not
    projected, they are assumed to be already projected.
--  ``.retr(u, t)`` – retraction map following vector ``u`` for time
+-  ``.retr(u, t=1.)`` – retraction map following vector ``u`` for time
    ``t``
--  ``.transp(u, t, v, *more)`` – transport vector ``v`` (and possibly
+-  ``.transp(v, *more, u, t=1.)`` – transport vector ``v`` (and possibly
    more vectors) with direction ``u`` for time ``t``
--  ``.retr_transp(u, t, v, *more)`` – transport ``self``, vector ``v``
+-  ``.retr_transp(v, *more, u, t=1.)`` – transport ``self``, vector ``v``
    (and possibly more vectors) with direction ``u`` for time ``t``
    (returns are plain tensors)
 

docs/devguide.rst

@@ -0,0 +1,18 @@
+Extending ``geoopt``
+====================
+
+Base Manifold
+-------------
+
+The common base class for all manifolds is :class:`geoopt.manifolds.base.Manifold`.
+
+.. autoclass:: geoopt.manifolds.base.Manifold
+    :private-members:
+    :members:
+
+
+Metaclass
+---------
+
+.. autoclass:: geoopt.manifolds.base.ManifoldMeta
+

docs/index.rst

@@ -17,11 +17,11 @@ API
    optimizers
    tensors
    samplers
+   devguide
 
 Indices and tables
 ==================
 
 * :ref:`genindex`
 * :ref:`modindex`
 * :ref:`search`
-

docs/manifolds.rst

@@ -3,12 +3,14 @@ Manifolds
 
 .. currentmodule:: geoopt.manifolds
 
-.. automodule:: geoopt.manifolds
-   :members: Euclidean, Stiefel, Sphere, SphereSubspaceComplementIntersection, SphereSubspaceIntersection
 
-Extending ``geoopt``
---------------------
-The common base class for all manifolds is :class:`geoopt.manifolds.base.Manifold`.
+All manifolds share same API. In order not to duplicate the same information, the complete public API is provided only for :class:`geoopt.manifolds.Euclidean` in the end of this file.
 
 .. automodule:: geoopt.manifolds
-   :members: Manifold
+    :members: Stiefel, Sphere, SphereSubspaceComplementIntersection, SphereSubspaceIntersection
+
+
+.. autoclass:: geoopt.manifolds.Euclidean
+    :members:
+    :inherited-members:
+

geoopt/linalg/_expm.py

@@ -7,7 +7,7 @@
 
 
 @torch.jit.script
-def torch_pade13(A):
+def torch_pade13(A):  # pragma: no cover
     # avoid torch select operation and unpack coefs
     (b0, b1, b2, b3, b4, b5, b6, b7, b8, b9, b10, b11, b12, b13) = (
         64764752532480000.0,
@@ -49,14 +49,14 @@ def torch_pade13(A):
 
 
 @torch.jit.script
-def matrix_2_power(x, p):
+def matrix_2_power(x, p):  # pragma: no cover
     for _ in range(int(p)):
         x = x @ x
     return x
 
 
 @torch.jit.script
-def expm_one(A):
+def expm_one(A):  # pragma: no cover
     # no checks, this is private implementation
     # but A should be a matrix
     A_fro = torch.norm(A)

geoopt/linalg/batch_linalg.py

@@ -5,7 +5,7 @@
 
 
 @torch.jit.script
-def svd(x):
+def svd(x):  # pragma: no cover
     # inspired by
     # https://discuss.pytorch.org/t/multidimensional-svd/4366/2
     # prolonged here:
@@ -32,7 +32,7 @@ def svd(x):
 
 
 @torch.jit.script
-def qr(x):
+def qr(x):  # pragma: no cover
     # inspired by
     # https://discuss.pytorch.org/t/multidimensional-svd/4366/2
     # prolonged here:
@@ -56,12 +56,12 @@ def qr(x):
 
 
 @torch.jit.script
-def sym(x):
+def sym(x):  # pragma: no cover
     return 0.5 * (x.transpose(-1, -2) + x)
 
 
 @torch.jit.script
-def extract_diag(x):
+def extract_diag(x):  # pragma: no cover
     n, m = x.shape[-2:]
     batch = x.shape[:-2]
     k = n if n < m else m
@@ -72,7 +72,7 @@ def extract_diag(x):
 
 
 @torch.jit.script
-def matrix_rank(x):
+def matrix_rank(x):  # pragma: no cover
     # inspired by
     # https://discuss.pytorch.org/t/multidimensional-svd/4366/2
     # prolonged here:
@@ -97,7 +97,7 @@ def matrix_rank(x):
 
 
 @torch.jit.script
-def expm(x):
+def expm(x):  # pragma: no cover
     # inspired by
     # https://discuss.pytorch.org/t/multidimensional-svd/4366/2
     # prolonged here: