[LLVM] Fix Maintainers.md formatting (NFC)
[llvm-project.git] / mlir / lib / Bindings / Python / DialectSparseTensor.cpp
bloba730bf500be98cf4c1e8049646725f8fc3e198b3
1 //===- DialectSparseTensor.cpp - 'sparse_tensor' dialect submodule --------===//
2 //
3 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
4 // See https://llvm.org/LICENSE.txt for license information.
5 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
6 //
7 //===----------------------------------------------------------------------===//
9 #include "mlir-c/AffineMap.h"
10 #include "mlir-c/Dialect/SparseTensor.h"
11 #include "mlir-c/IR.h"
12 #include "mlir/Bindings/Python/PybindAdaptors.h"
13 #include <optional>
14 #include <pybind11/cast.h>
15 #include <pybind11/detail/common.h>
16 #include <pybind11/pybind11.h>
17 #include <pybind11/pytypes.h>
18 #include <vector>
20 namespace py = pybind11;
21 using namespace llvm;
22 using namespace mlir;
23 using namespace mlir::python::adaptors;
25 static void populateDialectSparseTensorSubmodule(const py::module &m) {
26 py::enum_<MlirSparseTensorLevelFormat>(m, "LevelFormat", py::module_local())
27 .value("dense", MLIR_SPARSE_TENSOR_LEVEL_DENSE)
28 .value("n_out_of_m", MLIR_SPARSE_TENSOR_LEVEL_N_OUT_OF_M)
29 .value("compressed", MLIR_SPARSE_TENSOR_LEVEL_COMPRESSED)
30 .value("singleton", MLIR_SPARSE_TENSOR_LEVEL_SINGLETON)
31 .value("loose_compressed", MLIR_SPARSE_TENSOR_LEVEL_LOOSE_COMPRESSED);
33 py::enum_<MlirSparseTensorLevelPropertyNondefault>(m, "LevelProperty",
34 py::module_local())
35 .value("non_ordered", MLIR_SPARSE_PROPERTY_NON_ORDERED)
36 .value("non_unique", MLIR_SPARSE_PROPERTY_NON_UNIQUE)
37 .value("soa", MLIR_SPARSE_PROPERTY_SOA);
39 mlir_attribute_subclass(m, "EncodingAttr",
40 mlirAttributeIsASparseTensorEncodingAttr)
41 .def_classmethod(
42 "get",
43 [](py::object cls, std::vector<MlirSparseTensorLevelType> lvlTypes,
44 std::optional<MlirAffineMap> dimToLvl,
45 std::optional<MlirAffineMap> lvlToDim, int posWidth, int crdWidth,
46 std::optional<MlirAttribute> explicitVal,
47 std::optional<MlirAttribute> implicitVal, MlirContext context) {
48 return cls(mlirSparseTensorEncodingAttrGet(
49 context, lvlTypes.size(), lvlTypes.data(),
50 dimToLvl ? *dimToLvl : MlirAffineMap{nullptr},
51 lvlToDim ? *lvlToDim : MlirAffineMap{nullptr}, posWidth,
52 crdWidth, explicitVal ? *explicitVal : MlirAttribute{nullptr},
53 implicitVal ? *implicitVal : MlirAttribute{nullptr}));
55 py::arg("cls"), py::arg("lvl_types"), py::arg("dim_to_lvl"),
56 py::arg("lvl_to_dim"), py::arg("pos_width"), py::arg("crd_width"),
57 py::arg("explicit_val") = py::none(),
58 py::arg("implicit_val") = py::none(), py::arg("context") = py::none(),
59 "Gets a sparse_tensor.encoding from parameters.")
60 .def_classmethod(
61 "build_level_type",
62 [](py::object cls, MlirSparseTensorLevelFormat lvlFmt,
63 const std::vector<MlirSparseTensorLevelPropertyNondefault>
64 &properties,
65 unsigned n, unsigned m) {
66 return mlirSparseTensorEncodingAttrBuildLvlType(
67 lvlFmt, properties.data(), properties.size(), n, m);
69 py::arg("cls"), py::arg("lvl_fmt"),
70 py::arg("properties") =
71 std::vector<MlirSparseTensorLevelPropertyNondefault>(),
72 py::arg("n") = 0, py::arg("m") = 0,
73 "Builds a sparse_tensor.encoding.level_type from parameters.")
74 .def_property_readonly(
75 "lvl_types",
76 [](MlirAttribute self) {
77 const int lvlRank = mlirSparseTensorEncodingGetLvlRank(self);
78 std::vector<MlirSparseTensorLevelType> ret;
79 ret.reserve(lvlRank);
80 for (int l = 0; l < lvlRank; ++l)
81 ret.push_back(mlirSparseTensorEncodingAttrGetLvlType(self, l));
82 return ret;
84 .def_property_readonly(
85 "dim_to_lvl",
86 [](MlirAttribute self) -> std::optional<MlirAffineMap> {
87 MlirAffineMap ret = mlirSparseTensorEncodingAttrGetDimToLvl(self);
88 if (mlirAffineMapIsNull(ret))
89 return {};
90 return ret;
92 .def_property_readonly(
93 "lvl_to_dim",
94 [](MlirAttribute self) -> std::optional<MlirAffineMap> {
95 MlirAffineMap ret = mlirSparseTensorEncodingAttrGetLvlToDim(self);
96 if (mlirAffineMapIsNull(ret))
97 return {};
98 return ret;
100 .def_property_readonly("pos_width",
101 mlirSparseTensorEncodingAttrGetPosWidth)
102 .def_property_readonly("crd_width",
103 mlirSparseTensorEncodingAttrGetCrdWidth)
104 .def_property_readonly(
105 "explicit_val",
106 [](MlirAttribute self) -> std::optional<MlirAttribute> {
107 MlirAttribute ret =
108 mlirSparseTensorEncodingAttrGetExplicitVal(self);
109 if (mlirAttributeIsNull(ret))
110 return {};
111 return ret;
113 .def_property_readonly(
114 "implicit_val",
115 [](MlirAttribute self) -> std::optional<MlirAttribute> {
116 MlirAttribute ret =
117 mlirSparseTensorEncodingAttrGetImplicitVal(self);
118 if (mlirAttributeIsNull(ret))
119 return {};
120 return ret;
122 .def_property_readonly(
123 "structured_n",
124 [](MlirAttribute self) -> unsigned {
125 const int lvlRank = mlirSparseTensorEncodingGetLvlRank(self);
126 return mlirSparseTensorEncodingAttrGetStructuredN(
127 mlirSparseTensorEncodingAttrGetLvlType(self, lvlRank - 1));
129 .def_property_readonly(
130 "structured_m",
131 [](MlirAttribute self) -> unsigned {
132 const int lvlRank = mlirSparseTensorEncodingGetLvlRank(self);
133 return mlirSparseTensorEncodingAttrGetStructuredM(
134 mlirSparseTensorEncodingAttrGetLvlType(self, lvlRank - 1));
136 .def_property_readonly("lvl_formats_enum", [](MlirAttribute self) {
137 const int lvlRank = mlirSparseTensorEncodingGetLvlRank(self);
138 std::vector<MlirSparseTensorLevelFormat> ret;
139 ret.reserve(lvlRank);
140 for (int l = 0; l < lvlRank; l++)
141 ret.push_back(mlirSparseTensorEncodingAttrGetLvlFmt(self, l));
142 return ret;
146 PYBIND11_MODULE(_mlirDialectsSparseTensor, m) {
147 m.doc() = "MLIR SparseTensor dialect.";
148 populateDialectSparseTensorSubmodule(m);