Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
Menu
Open sidebar
OpenDAS
dgl
Commits
ed9ef14a
Unverified
Commit
ed9ef14a
authored
Jan 18, 2024
by
Ramon Zhou
Committed by
GitHub
Jan 18, 2024
Browse files
[Misc] Turn off `fused` in DGL rgcn example (#6977)
parent
cf8fea73
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
2 additions
and
2 deletions
+2
-2
examples/core/rgcn/hetero_rgcn.py
examples/core/rgcn/hetero_rgcn.py
+2
-2
No files found.
examples/core/rgcn/hetero_rgcn.py
View file @
ed9ef14a
...
@@ -116,7 +116,7 @@ def prepare_data(args, device):
...
@@ -116,7 +116,7 @@ def prepare_data(args, device):
# Initialize a train sampler that samples neighbors for multi-layer graph
# Initialize a train sampler that samples neighbors for multi-layer graph
# convolution. It samples 25 and 10 neighbors for the first and second
# convolution. It samples 25 and 10 neighbors for the first and second
# layers respectively.
# layers respectively.
sampler
=
dgl
.
dataloading
.
MultiLayerNeighborSampler
([
25
,
10
])
sampler
=
dgl
.
dataloading
.
MultiLayerNeighborSampler
([
25
,
10
]
,
fused
=
False
)
num_workers
=
args
.
num_workers
num_workers
=
args
.
num_workers
train_loader
=
dgl
.
dataloading
.
DataLoader
(
train_loader
=
dgl
.
dataloading
.
DataLoader
(
g
,
g
,
...
@@ -488,7 +488,7 @@ def evaluate(
...
@@ -488,7 +488,7 @@ def evaluate(
else
:
else
:
evaluator
=
MAG240MEvaluator
()
evaluator
=
MAG240MEvaluator
()
sampler
=
dgl
.
dataloading
.
MultiLayerNeighborSampler
([
25
,
10
])
sampler
=
dgl
.
dataloading
.
MultiLayerNeighborSampler
([
25
,
10
]
,
fused
=
False
)
dataloader
=
dgl
.
dataloading
.
DataLoader
(
dataloader
=
dgl
.
dataloading
.
DataLoader
(
g
,
g
,
idx
,
idx
,
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment