forked from pytorch/pytorch
-
Notifications
You must be signed in to change notification settings - Fork 0
/
_lowrank.py
281 lines (221 loc) · 10.7 KB
/
_lowrank.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
"""Implement various linear algebra algorithms for low rank matrices.
"""
__all__ = ['svd_lowrank', 'pca_lowrank']
from torch import Tensor
import torch
from . import _linalg_utils as _utils
from .overrides import has_torch_function, handle_torch_function
from typing import Optional, Tuple
def get_approximate_basis(A: Tensor,
q: int,
niter: Optional[int] = 2,
M: Optional[Tensor] = None
) -> Tensor:
"""Return tensor :math:`Q` with :math:`q` orthonormal columns such
that :math:`Q Q^H A` approximates :math:`A`. If :math:`M` is
specified, then :math:`Q` is such that :math:`Q Q^H (A - M)`
approximates :math:`A - M`.
.. note:: The implementation is based on the Algorithm 4.4 from
Halko et al, 2009.
.. note:: For an adequate approximation of a k-rank matrix
:math:`A`, where k is not known in advance but could be
estimated, the number of :math:`Q` columns, q, can be
choosen according to the following criteria: in general,
:math:`k <= q <= min(2*k, m, n)`. For large low-rank
matrices, take :math:`q = k + 5..10`. If k is
relatively small compared to :math:`min(m, n)`, choosing
:math:`q = k + 0..2` may be sufficient.
.. note:: To obtain repeatable results, reset the seed for the
pseudorandom number generator
Args::
A (Tensor): the input tensor of size :math:`(*, m, n)`
q (int): the dimension of subspace spanned by :math:`Q`
columns.
niter (int, optional): the number of subspace iterations to
conduct; ``niter`` must be a
nonnegative integer. In most cases, the
default value 2 is more than enough.
M (Tensor, optional): the input tensor's mean of size
:math:`(*, 1, n)`.
References::
- Nathan Halko, Per-Gunnar Martinsson, and Joel Tropp, Finding
structure with randomness: probabilistic algorithms for
constructing approximate matrix decompositions,
arXiv:0909.4061 [math.NA; math.PR], 2009 (available at
`arXiv <http://arxiv.org/abs/0909.4061>`_).
"""
niter = 2 if niter is None else niter
m, n = A.shape[-2:]
dtype = _utils.get_floating_dtype(A)
matmul = _utils.matmul
R = torch.randn(n, q, dtype=dtype, device=A.device)
# The following code could be made faster using torch.geqrf + torch.ormqr
# but geqrf is not differentiable
A_H = _utils.transjugate(A)
if M is None:
Q = torch.linalg.qr(matmul(A, R)).Q
for i in range(niter):
Q = torch.linalg.qr(matmul(A_H, Q)).Q
Q = torch.linalg.qr(matmul(A, Q)).Q
else:
M_H = _utils.transjugate(M)
Q = torch.linalg.qr(matmul(A, R) - matmul(M, R)).Q
for i in range(niter):
Q = torch.linalg.qr(matmul(A_H, Q) - matmul(M_H, Q)).Q
Q = torch.linalg.qr(matmul(A, Q) - matmul(M, Q)).Q
return Q
def svd_lowrank(A: Tensor, q: Optional[int] = 6, niter: Optional[int] = 2,
M: Optional[Tensor] = None) -> Tuple[Tensor, Tensor, Tensor]:
r"""Return the singular value decomposition ``(U, S, V)`` of a matrix,
batches of matrices, or a sparse matrix :math:`A` such that
:math:`A \approx U diag(S) V^T`. In case :math:`M` is given, then
SVD is computed for the matrix :math:`A - M`.
.. note:: The implementation is based on the Algorithm 5.1 from
Halko et al, 2009.
.. note:: To obtain repeatable results, reset the seed for the
pseudorandom number generator
.. note:: The input is assumed to be a low-rank matrix.
.. note:: In general, use the full-rank SVD implementation
:func:`torch.linalg.svd` for dense matrices due to its 10-fold
higher performance characteristics. The low-rank SVD
will be useful for huge sparse matrices that
:func:`torch.linalg.svd` cannot handle.
Args::
A (Tensor): the input tensor of size :math:`(*, m, n)`
q (int, optional): a slightly overestimated rank of A.
niter (int, optional): the number of subspace iterations to
conduct; niter must be a nonnegative
integer, and defaults to 2
M (Tensor, optional): the input tensor's mean of size
:math:`(*, 1, n)`.
References::
- Nathan Halko, Per-Gunnar Martinsson, and Joel Tropp, Finding
structure with randomness: probabilistic algorithms for
constructing approximate matrix decompositions,
arXiv:0909.4061 [math.NA; math.PR], 2009 (available at
`arXiv <https://arxiv.org/abs/0909.4061>`_).
"""
if not torch.jit.is_scripting():
tensor_ops = (A, M)
if (not set(map(type, tensor_ops)).issubset((torch.Tensor, type(None))) and has_torch_function(tensor_ops)):
return handle_torch_function(svd_lowrank, tensor_ops, A, q=q, niter=niter, M=M)
return _svd_lowrank(A, q=q, niter=niter, M=M)
def _svd_lowrank(A: Tensor, q: Optional[int] = 6, niter: Optional[int] = 2,
M: Optional[Tensor] = None) -> Tuple[Tensor, Tensor, Tensor]:
q = 6 if q is None else q
m, n = A.shape[-2:]
matmul = _utils.matmul
if M is None:
M_t = None
else:
M_t = _utils.transpose(M)
A_t = _utils.transpose(A)
# Algorithm 5.1 in Halko et al 2009, slightly modified to reduce
# the number conjugate and transpose operations
if m < n or n > q:
# computing the SVD approximation of a transpose in
# order to keep B shape minimal (the m < n case) or the V
# shape small (the n > q case)
Q = get_approximate_basis(A_t, q, niter=niter, M=M_t)
Q_c = _utils.conjugate(Q)
if M is None:
B_t = matmul(A, Q_c)
else:
B_t = matmul(A, Q_c) - matmul(M, Q_c)
assert B_t.shape[-2] == m, (B_t.shape, m)
assert B_t.shape[-1] == q, (B_t.shape, q)
assert B_t.shape[-1] <= B_t.shape[-2], B_t.shape
U, S, Vh = torch.linalg.svd(B_t, full_matrices=False)
V = Vh.mH
V = Q.matmul(V)
else:
Q = get_approximate_basis(A, q, niter=niter, M=M)
Q_c = _utils.conjugate(Q)
if M is None:
B = matmul(A_t, Q_c)
else:
B = matmul(A_t, Q_c) - matmul(M_t, Q_c)
B_t = _utils.transpose(B)
assert B_t.shape[-2] == q, (B_t.shape, q)
assert B_t.shape[-1] == n, (B_t.shape, n)
assert B_t.shape[-1] <= B_t.shape[-2], B_t.shape
U, S, Vh = torch.linalg.svd(B_t, full_matrices=False)
V = Vh.mH
U = Q.matmul(U)
return U, S, V
def pca_lowrank(A: Tensor, q: Optional[int] = None, center: bool = True,
niter: int = 2) -> Tuple[Tensor, Tensor, Tensor]:
r"""Performs linear Principal Component Analysis (PCA) on a low-rank
matrix, batches of such matrices, or sparse matrix.
This function returns a namedtuple ``(U, S, V)`` which is the
nearly optimal approximation of a singular value decomposition of
a centered matrix :math:`A` such that :math:`A = U diag(S) V^T`.
.. note:: The relation of ``(U, S, V)`` to PCA is as follows:
- :math:`A` is a data matrix with ``m`` samples and
``n`` features
- the :math:`V` columns represent the principal directions
- :math:`S ** 2 / (m - 1)` contains the eigenvalues of
:math:`A^T A / (m - 1)` which is the covariance of
``A`` when ``center=True`` is provided.
- ``matmul(A, V[:, :k])`` projects data to the first k
principal components
.. note:: Different from the standard SVD, the size of returned
matrices depend on the specified rank and q
values as follows:
- :math:`U` is m x q matrix
- :math:`S` is q-vector
- :math:`V` is n x q matrix
.. note:: To obtain repeatable results, reset the seed for the
pseudorandom number generator
Args:
A (Tensor): the input tensor of size :math:`(*, m, n)`
q (int, optional): a slightly overestimated rank of
:math:`A`. By default, ``q = min(6, m,
n)``.
center (bool, optional): if True, center the input tensor,
otherwise, assume that the input is
centered.
niter (int, optional): the number of subspace iterations to
conduct; niter must be a nonnegative
integer, and defaults to 2.
References::
- Nathan Halko, Per-Gunnar Martinsson, and Joel Tropp, Finding
structure with randomness: probabilistic algorithms for
constructing approximate matrix decompositions,
arXiv:0909.4061 [math.NA; math.PR], 2009 (available at
`arXiv <http://arxiv.org/abs/0909.4061>`_).
"""
if not torch.jit.is_scripting():
if type(A) is not torch.Tensor and has_torch_function((A,)):
return handle_torch_function(pca_lowrank, (A,), A, q=q, center=center, niter=niter)
(m, n) = A.shape[-2:]
if q is None:
q = min(6, m, n)
elif not (q >= 0 and q <= min(m, n)):
raise ValueError('q(={}) must be non-negative integer'
' and not greater than min(m, n)={}'
.format(q, min(m, n)))
if not (niter >= 0):
raise ValueError('niter(={}) must be non-negative integer'
.format(niter))
dtype = _utils.get_floating_dtype(A)
if not center:
return _svd_lowrank(A, q, niter=niter, M=None)
if _utils.is_sparse(A):
if len(A.shape) != 2:
raise ValueError('pca_lowrank input is expected to be 2-dimensional tensor')
c = torch.sparse.sum(A, dim=(-2,)) / m
# reshape c
column_indices = c.indices()[0]
indices = torch.zeros(2, len(column_indices),
dtype=column_indices.dtype,
device=column_indices.device)
indices[0] = column_indices
C_t = torch.sparse_coo_tensor(
indices, c.values(), (n, 1), dtype=dtype, device=A.device)
ones_m1_t = torch.ones(A.shape[:-2] + (1, m), dtype=dtype, device=A.device)
M = _utils.transpose(torch.sparse.mm(C_t, ones_m1_t))
return _svd_lowrank(A, q, niter=niter, M=M)
else:
C = A.mean(dim=(-2,), keepdim=True)
return _svd_lowrank(A - C, q, niter=niter, M=None)