Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[Optimization] Use scipy's eigs instead of numpy in lap_pe #5855

Merged
merged 8 commits into from
Aug 24, 2023
Merged

[Optimization] Use scipy's eigs instead of numpy in lap_pe #5855

Changes from 6 commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
7eeaceb
use scipy instead of numpy
BarclayII Jun 12, 2023
5b62d1b
Update functional.py
BarclayII Jun 29, 2023
5b4f135
fix
BarclayII Jul 4, 2023
19317d4
Update functional.py
BarclayII Jul 5, 2023
c40c8a2
Update python/dgl/transforms/functional.py
rudongyu Jul 27, 2023
5b07242
lint
BarclayII Aug 3, 2023
bc6a514
Update python/dgl/transforms/functional.py
BarclayII Aug 21, 2023
070f0fd
Update python/dgl/transforms/functional.py
BarclayII Aug 21, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
26 changes: 19 additions & 7 deletions python/dgl/transforms/functional.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3672,13 +3672,25 @@ def lap_pe(g, k, padding=False, return_eigval=False):
L = sparse.eye(g.num_nodes()) - N * A * N

# select eigenvectors with smaller eigenvalues O(n + klogk)
EigVal, EigVec = np.linalg.eig(L.toarray())
max_freqs = min(n - 1, k)
kpartition_indices = np.argpartition(EigVal, max_freqs)[: max_freqs + 1]
topk_eigvals = EigVal[kpartition_indices]
topk_indices = kpartition_indices[topk_eigvals.argsort()][1:]
topk_EigVec = EigVec[:, topk_indices]
eigvals = F.tensor(EigVal[topk_indices], dtype=F.float32)

BarclayII marked this conversation as resolved.
Show resolved Hide resolved
if k + 1 < n - 1:
# Use scipy if k + 1 < n - 1 for memory efficiency.
EigVal, EigVec = scipy.sparse.linalg.eigs(
L, k=k + 1, which="SR", tol=1e-2
)
topk_indices = EigVal.argsort()[1:]
topk_eigvals = EigVal[topk_indices].real # scipy may give complex
BarclayII marked this conversation as resolved.
Show resolved Hide resolved
BarclayII marked this conversation as resolved.
Show resolved Hide resolved
topk_EigVec = EigVec[:, topk_indices].real
else:
# Fallback to numpy since scipy.sparse do not support this case.
EigVal, EigVec = np.linalg.eig(L.toarray())
max_freqs = min(n - 1, k)
kpartition_indices = np.argpartition(EigVal, max_freqs)[: max_freqs + 1]
topk_eigvals = EigVal[kpartition_indices]
topk_indices = kpartition_indices[topk_eigvals.argsort()][1:]
topk_EigVec = EigVec[:, topk_indices]
topk_EigVal = EigVal[topk_indices]
eigvals = F.tensor(topk_EigVal, dtype=F.float32)

# get random flip signs
rand_sign = 2 * (np.random.rand(max_freqs) > 0.5) - 1.0
Expand Down