File size: 9,013 Bytes
c3d0544
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
# SPDX-FileCopyrightText: Copyright (c) 2023 - 2025 NVIDIA CORPORATION & AFFILIATES.
# SPDX-FileCopyrightText: All rights reserved.
# SPDX-License-Identifier: Apache-2.0
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

"""Graph backend for creating DGL or PyG graphs."""

from types import NoneType
from typing import List, Optional, Tuple, TypeAlias, Union

import torch
from torch import Tensor, testing

try:
    from dgl import DGLGraph

    DGL_AVAILABLE = True
except ImportError:
    DGL_AVAILABLE = False
    DGLGraph: TypeAlias = NoneType

try:
    import torch_geometric.utils as pyg_utils
    from torch_geometric.data import Data as PyGData
    from torch_geometric.data import HeteroData as PyGHeteroData

    PYG_AVAILABLE = True
except ImportError:
    PYG_AVAILABLE = False
    PyGData: TypeAlias = NoneType

from physicsnemo.models.gnn_layers.utils import GraphType
from physicsnemo.utils.graphcast.graph_utils import (
    azimuthal_angle,
    geospatial_rotation,
    polar_angle,
    xyz2latlon,
)


class DglGraphBackend:
    """DGL graph backend."""

    name: str = "dgl"

    @staticmethod
    def create_graph(
        src: List,
        dst: List,
        to_bidirected: bool,
        add_self_loop: bool,
        dtype: torch.dtype,
    ) -> DGLGraph:
        """Create DGL graph."""
        from physicsnemo.utils.graphcast.graph_utils_dgl import create_graph

        return create_graph(src, dst, to_bidirected, add_self_loop, dtype)

    @staticmethod
    def create_heterograph(
        src: List,
        dst: List,
        labels: str,
        dtype: torch.dtype = torch.int32,
        num_nodes_dict: Optional[dict] = None,
    ) -> DGLGraph:
        """Create heterogeneous graph using DGL."""
        from physicsnemo.utils.graphcast.graph_utils_dgl import create_heterograph

        return create_heterograph(src, dst, labels, dtype, num_nodes_dict)

    @staticmethod
    def add_edge_features(
        graph: DGLGraph, pos: Tensor, normalize: bool = True
    ) -> DGLGraph:
        """Add edge features to DGL graph."""
        from physicsnemo.utils.graphcast.graph_utils_dgl import add_edge_features

        return add_edge_features(graph, pos, normalize)

    @staticmethod
    def add_node_features(graph: DGLGraph, pos: Tensor) -> DGLGraph:
        """Add node features to DGL graph."""
        from physicsnemo.utils.graphcast.graph_utils_dgl import add_node_features

        return add_node_features(graph, pos)

    @staticmethod
    def khop_adj_all_k(graph: DGLGraph, kmax: int):
        """Construct the union of k-hop adjacencies up to distance `kmax` for a graph."""

        if not graph.is_homogeneous:
            raise NotImplementedError("only homogeneous graph is supported")
        min_degree = graph.in_degrees().min()
        with torch.no_grad():
            adj = graph.adj_external(transpose=True, scipy_fmt=None)
            adj_k = adj
            adj_all = adj.clone()
            for _ in range(2, kmax + 1):
                # scale with min-degree to avoid too large values
                # but >= 1.0
                adj_k = (adj @ adj_k) / min_degree
                adj_all += adj_k
        return adj_all.to_dense().bool()


class PyGGraphBackend:
    """PyG graph backend."""

    name: str = "pyg"

    @staticmethod
    def create_graph(
        src: List,
        dst: List,
        to_bidirected: bool,
        add_self_loop: bool,
        dtype: torch.dtype = torch.int64,
    ) -> PyGData:
        """Create PyG graph.

        dtype is ignored for PyG graph backend since PyG only supports int64 dtype.
        """

        edge_index = torch.stack([torch.tensor(src), torch.tensor(dst)], dim=0).long()
        if to_bidirected:
            edge_index = pyg_utils.to_undirected(edge_index)
        if add_self_loop:
            edge_index, _ = pyg_utils.add_self_loops(edge_index)

        return PyGData(edge_index=edge_index)

    @staticmethod
    def create_heterograph(
        src: List,
        dst: List,
        labels: str,
        dtype: torch.dtype = torch.int64,
    ) -> GraphType:
        """Create heterogeneous graph using PyG.

        Parameters
        ----------
        src : List
            List of source nodes
        dst : List
            List of destination nodes
        labels : str
            Label of the edge type
        dtype : torch.dtype, optional
            Graph index data type, ignored for PyG graph backend since PyG only supports int64 dtype.

        Returns
        -------
        GraphType
            Heterogeneous graph object
        """

        g = PyGHeteroData()
        g[labels].edge_index = torch.stack(
            [torch.tensor(src), torch.tensor(dst)], dim=0
        ).long()

        return g

    @staticmethod
    def add_edge_features(
        graph: PyGData,
        pos: Union[Tensor, Tuple[Tensor, Tensor]],
        normalize: bool = True,
    ) -> PyGData:
        """Add edge features to PyG graph."""

        if isinstance(pos, tuple):
            src_pos, dst_pos = pos
        else:
            src_pos = dst_pos = pos

        if isinstance(graph, PyGData):
            src, dst = graph.edge_index
        elif isinstance(graph, PyGHeteroData):
            src, dst = graph[graph.edge_types[0]].edge_index
        else:
            raise ValueError(f"Invalid graph type: {type(graph)}")

        src_pos, dst_pos = src_pos[src.long()], dst_pos[dst.long()]
        dst_latlon = xyz2latlon(dst_pos, unit="rad")
        dst_lat, dst_lon = dst_latlon[:, 0], dst_latlon[:, 1]

        # Azimuthal & polar rotation (same logic as DGL version)
        theta_azimuthal = azimuthal_angle(dst_lon)
        theta_polar = polar_angle(dst_lat)

        src_pos = geospatial_rotation(
            src_pos, theta=theta_azimuthal, axis="z", unit="rad"
        )
        dst_pos = geospatial_rotation(
            dst_pos, theta=theta_azimuthal, axis="z", unit="rad"
        )

        # Validation checks
        try:
            testing.assert_close(dst_pos[:, 1], torch.zeros_like(dst_pos[:, 1]))
        except ValueError:
            raise ValueError(
                "Invalid projection of edge nodes to local coordinate system"
            )

        src_pos = geospatial_rotation(src_pos, theta=theta_polar, axis="y", unit="rad")
        dst_pos = geospatial_rotation(dst_pos, theta=theta_polar, axis="y", unit="rad")

        # More validation checks
        try:
            testing.assert_close(dst_pos[:, 0], torch.ones_like(dst_pos[:, 0]))
            testing.assert_close(dst_pos[:, 1], torch.zeros_like(dst_pos[:, 1]))
            testing.assert_close(dst_pos[:, 2], torch.zeros_like(dst_pos[:, 2]))
        except ValueError:
            raise ValueError(
                "Invalid projection of edge nodes to local coordinate system"
            )

        # Prepare edge features
        disp = src_pos - dst_pos
        disp_norm = torch.linalg.norm(disp, dim=-1, keepdim=True)

        if normalize:
            max_disp_norm = torch.max(disp_norm)
            graph.edge_attr = torch.cat(
                (disp / max_disp_norm, disp_norm / max_disp_norm), dim=-1
            )
        else:
            graph.edge_attr = torch.cat((disp, disp_norm), dim=-1)

        return graph

    @staticmethod
    def add_node_features(graph: PyGData, pos: Tensor) -> PyGData:
        """Add node features to PyG graph."""

        latlon = xyz2latlon(pos)
        lat, lon = latlon[:, 0], latlon[:, 1]
        graph.x = torch.stack((torch.cos(lat), torch.sin(lon), torch.cos(lon)), dim=-1)
        return graph

    @staticmethod
    def khop_adj_all_k(graph: PyGData, kmax: int):
        """Construct the union of k-hop adjacencies up to distance `kmax` for a graph."""

        from torch_sparse import SparseTensor

        if not isinstance(graph, PyGData):
            raise ValueError(
                f"Invalid graph type: {type(graph)}, only Data type is supported."
            )

        if graph.edge_index is None:
            raise ValueError("Graph must have edge_index defined.")

        n_nodes = graph.num_nodes

        # Build SparseTensor adjacency: shape [n_nodes, n_nodes]
        # row = source, col = target
        adj = SparseTensor.from_edge_index(
            graph.edge_index, sparse_sizes=(n_nodes, n_nodes)
        )

        adj_k = adj.clone()
        adj_all = adj.clone()

        for _ in range(2, kmax + 1):
            adj_k = adj @ adj_k
            adj_all = adj_all + adj_k

        return adj_all.to_dense().bool()