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
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
|
// Copyright (C) 2017-2019 Chris Richardson and Garth N. Wells
//
// This file is part of DOLFINx (https://www.fenicsproject.org)
//
// SPDX-License-Identifier: LGPL-3.0-or-later
#include "dolfinx_wrappers/array.h"
#include "dolfinx_wrappers/caster_mpi.h"
#include "dolfinx_wrappers/numpy_dtype.h"
#include <complex>
#include <cstdint>
#include <dolfinx/common/IndexMap.h>
#include <dolfinx/la/MatrixCSR.h>
#include <dolfinx/la/SparsityPattern.h>
#include <dolfinx/la/Vector.h>
#include <dolfinx/la/utils.h>
#include <memory>
#include <nanobind/nanobind.h>
#include <nanobind/ndarray.h>
#include <nanobind/stl/array.h>
#include <nanobind/stl/complex.h>
#include <nanobind/stl/pair.h>
#include <nanobind/stl/shared_ptr.h>
#include <nanobind/stl/vector.h>
#include <span>
namespace nb = nanobind;
using namespace nb::literals;
namespace
{
// InsertMode types
enum class PyInsertMode : std::uint8_t
{
add,
insert
};
// Declare objects that have multiple scalar types
template <typename T>
void declare_objects(nb::module_& m, const std::string& type)
{
// dolfinx::la::Vector
std::string pyclass_vector_name = std::string("Vector_") + type;
nb::class_<dolfinx::la::Vector<T>>(m, pyclass_vector_name.c_str())
.def(nb::init<std::shared_ptr<const dolfinx::common::IndexMap>, int>(),
nb::arg("map"), nb::arg("bs"))
.def(nb::init<const dolfinx::la::Vector<T>&>(), nb::arg("vec"))
.def_prop_ro("dtype", [](const dolfinx::la::Vector<T>&)
{ return dolfinx_wrappers::numpy_dtype<T>(); })
.def_prop_ro("index_map", &dolfinx::la::Vector<T>::index_map)
.def_prop_ro("bs", &dolfinx::la::Vector<T>::bs)
.def_prop_ro(
"array",
[](dolfinx::la::Vector<T>& self)
{
return nb::ndarray<T, nb::numpy>(self.array().data(),
{self.array().size()});
},
nb::rv_policy::reference_internal)
.def("scatter_forward",
[](dolfinx::la::Vector<T>& self) { self.scatter_fwd(); })
.def(
"scatter_reverse",
[](dolfinx::la::Vector<T>& self, PyInsertMode mode)
{
switch (mode)
{
case PyInsertMode::add: // Add
self.scatter_rev(std::plus<T>());
break;
case PyInsertMode::insert: // Insert
self.scatter_rev([](T /*a*/, T b) { return b; });
break;
default:
throw std::runtime_error("InsertMode not recognized.");
break;
}
},
nb::arg("mode"));
// dolfinx::la::MatrixCSR
std::string pyclass_matrix_name = std::string("MatrixCSR_") + type;
nb::class_<dolfinx::la::MatrixCSR<
T, std::vector<T>, std::vector<std::int32_t>, std::vector<std::int64_t>>>(
m, pyclass_matrix_name.c_str())
.def(nb::init<const dolfinx::la::SparsityPattern&,
dolfinx::la::BlockMode>(),
nb::arg("p"),
nb::arg("block_mode") = dolfinx::la::BlockMode::compact)
.def_prop_ro("dtype", [](const dolfinx::la::MatrixCSR<T>&)
{ return dolfinx_wrappers::numpy_dtype<T>(); })
.def_prop_ro("bs", &dolfinx::la::MatrixCSR<T>::block_size)
.def("squared_norm", &dolfinx::la::MatrixCSR<T>::squared_norm)
.def("index_map", &dolfinx::la::MatrixCSR<T>::index_map)
.def("add",
[](dolfinx::la::MatrixCSR<T>& self, const std::vector<T>& x,
const std::vector<std::int32_t>& rows,
const std::vector<std::int32_t>& cols, int bs = 1)
{
if (bs == 1)
self.template add<1, 1>(x, rows, cols);
else if (bs == 2)
self.template add<2, 2>(x, rows, cols);
else if (bs == 3)
self.template add<3, 3>(x, rows, cols);
else
{
throw std::runtime_error(
"Block size not supported in this function");
}
})
.def("set",
[](dolfinx::la::MatrixCSR<T>& self, const std::vector<T>& x,
const std::vector<std::int32_t>& rows,
const std::vector<std::int32_t>& cols, int bs = 1)
{
if (bs == 1)
self.template set<1, 1>(x, rows, cols);
else if (bs == 2)
self.template set<2, 2>(x, rows, cols);
else if (bs == 3)
self.template set<3, 3>(x, rows, cols);
else
{
throw std::runtime_error(
"Block size not supported in this function");
}
})
.def("scatter_reverse", &dolfinx::la::MatrixCSR<T>::scatter_rev)
.def("mult", &dolfinx::la::MatrixCSR<T>::mult)
.def("to_dense",
[](const dolfinx::la::MatrixCSR<T>& self)
{
std::array<int, 2> bs = self.block_size();
std::size_t nrows = self.num_all_rows() * bs[0];
std::size_t ncols = self.index_map(1)->size_global() * bs[1];
std::vector<T> dense = self.to_dense();
assert(nrows * ncols == dense.size());
return dolfinx_wrappers::as_nbarray(std::move(dense),
{nrows, ncols});
})
.def_prop_ro(
"data",
[](dolfinx::la::MatrixCSR<T>& self)
{
return nb::ndarray<T, nb::numpy>(self.values().data(),
{self.values().size()});
},
nb::rv_policy::reference_internal)
.def_prop_ro(
"indices",
[](dolfinx::la::MatrixCSR<T>& self)
{
return nb::ndarray<const std::int32_t, nb::numpy>(
self.cols().data(), {self.cols().size()});
},
nb::rv_policy::reference_internal)
.def_prop_ro(
"indptr",
[](dolfinx::la::MatrixCSR<T>& self)
{
return nb::ndarray<const std::int64_t, nb::numpy>(
self.row_ptr().data(), {self.row_ptr().size()});
},
nb::rv_policy::reference_internal)
.def("scatter_rev_begin", &dolfinx::la::MatrixCSR<T>::scatter_rev_begin)
.def("scatter_rev_end", &dolfinx::la::MatrixCSR<T>::scatter_rev_end);
}
// Declare objects that have multiple scalar types
template <typename T>
void declare_functions(nb::module_& m)
{
m.def(
"norm", [](const dolfinx::la::Vector<T>& x, dolfinx::la::Norm type)
{ return dolfinx::la::norm(x, type); }, "vector"_a, "type"_a);
m.def(
"inner_product",
[](const dolfinx::la::Vector<T>& x, const dolfinx::la::Vector<T>& y)
{ return dolfinx::la::inner_product(x, y); }, nb::arg("x"), nb::arg("y"));
m.def(
"orthonormalize",
[](std::vector<dolfinx::la::Vector<T>*> basis)
{
std::vector<std::reference_wrapper<dolfinx::la::Vector<T>>> _basis;
_basis.reserve(basis.size());
for (std::size_t i = 0; i < basis.size(); ++i)
_basis.push_back(*basis[i]);
dolfinx::la::orthonormalize(_basis);
},
nb::arg("basis"));
m.def(
"is_orthonormal",
[](std::vector<const dolfinx::la::Vector<T>*> basis,
dolfinx::scalar_value_t<T> eps)
{
std::vector<std::reference_wrapper<const dolfinx::la::Vector<T>>>
_basis;
_basis.reserve(basis.size());
for (std::size_t i = 0; i < basis.size(); ++i)
_basis.push_back(*basis[i]);
return dolfinx::la::is_orthonormal(_basis, eps);
},
nb::arg("basis"), nb::arg("eps"));
}
} // namespace
namespace dolfinx_wrappers
{
void la(nb::module_& m)
{
nb::enum_<PyInsertMode>(m, "InsertMode")
.value("add", PyInsertMode::add)
.value("insert", PyInsertMode::insert);
nb::enum_<dolfinx::la::BlockMode>(m, "BlockMode")
.value("compact", dolfinx::la::BlockMode::compact)
.value("expanded", dolfinx::la::BlockMode::expanded);
nb::enum_<dolfinx::la::Norm>(m, "Norm")
.value("l1", dolfinx::la::Norm::l1, "l1 norm")
.value("l2", dolfinx::la::Norm::l2, "l2 norm")
.value("linf", dolfinx::la::Norm::linf, "linf norm")
.value("frobenius", dolfinx::la::Norm::frobenius, "Frobenius norm");
// dolfinx::la::SparsityPattern
nb::class_<dolfinx::la::SparsityPattern>(m, "SparsityPattern")
.def(
"__init__",
[](dolfinx::la::SparsityPattern* sp, MPICommWrapper comm,
std::array<std::shared_ptr<const dolfinx::common::IndexMap>, 2>
maps,
std::array<int, 2> bs)
{ new (sp) dolfinx::la::SparsityPattern(comm.get(), maps, bs); },
nb::arg("comm"), nb::arg("maps"), nb::arg("bs"))
.def(
"__init__",
[](dolfinx::la::SparsityPattern* sp, MPICommWrapper comm,
const std::vector<
std::vector<const dolfinx::la::SparsityPattern*>>& patterns,
const std::array<
std::vector<std::pair<
std::reference_wrapper<const dolfinx::common::IndexMap>,
int>>,
2>& maps,
std::array<std::vector<int>, 2> bs)
{
new (sp)
dolfinx::la::SparsityPattern(comm.get(), patterns, maps, bs);
},
nb::arg("comm"), nb::arg("patterns"), nb::arg("maps"), nb::arg("bs"))
.def("index_map", &dolfinx::la::SparsityPattern::index_map,
nb::arg("dim"))
.def("column_index_map", &dolfinx::la::SparsityPattern::column_index_map)
.def("finalize", &dolfinx::la::SparsityPattern::finalize)
.def_prop_ro("num_nonzeros", &dolfinx::la::SparsityPattern::num_nonzeros)
.def(
"insert",
[](dolfinx::la::SparsityPattern& self,
nb::ndarray<const std::int32_t, nb::ndim<1>, nb::c_contig> rows,
nb::ndarray<const std::int32_t, nb::ndim<1>, nb::c_contig> cols)
{
self.insert(std::span(rows.data(), rows.size()),
std::span(cols.data(), cols.size()));
},
nb::arg("rows"), nb::arg("cols"))
.def("insert",
nb::overload_cast<int32_t, int32_t>(
&dolfinx::la::SparsityPattern::insert),
nb::arg("row"), nb::arg("col"))
.def(
"insert_diagonal",
[](dolfinx::la::SparsityPattern& self,
nb::ndarray<const std::int32_t, nb::ndim<1>, nb::c_contig> rows)
{ self.insert_diagonal(std::span(rows.data(), rows.size())); },
nb::arg("rows"))
.def_prop_ro(
"graph",
[](dolfinx::la::SparsityPattern& self)
{
auto [edges, ptr] = self.graph();
return std::pair(nb::ndarray<const std::int32_t, nb::numpy>(
edges.data(), {edges.size()}),
nb::ndarray<const std::int64_t, nb::numpy>(
ptr.data(), {ptr.size()}));
},
nb::rv_policy::reference_internal);
// Declare objects that are templated over type
declare_objects<std::int8_t>(m, "int8");
declare_objects<std::int32_t>(m, "int32");
declare_objects<std::int64_t>(m, "int64");
declare_objects<float>(m, "float32");
declare_objects<double>(m, "float64");
declare_objects<std::complex<float>>(m, "complex64");
declare_objects<std::complex<double>>(m, "complex128");
declare_functions<float>(m);
declare_functions<double>(m);
declare_functions<std::complex<float>>(m);
declare_functions<std::complex<double>>(m);
}
} // namespace dolfinx_wrappers
|
