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
4fca8817
Unverified
Commit
4fca8817
authored
Oct 20, 2023
by
Rhett Ying
Committed by
GitHub
Oct 20, 2023
Browse files
[GraphBolt] update statistics with indexing enabled ItemSampler (#6478)
parent
557c0a86
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
5 additions
and
5 deletions
+5
-5
examples/sampling/graphbolt/rgcn/README.md
examples/sampling/graphbolt/rgcn/README.md
+5
-5
No files found.
examples/sampling/graphbolt/rgcn/README.md
View file @
4fca8817
...
...
@@ -11,7 +11,7 @@ python3 hetero_rgcn.py --dataset ogbn-mag
### Sample on CPU and train/infer on GPU
```
python3 hetero_rgcn.py --dataset ogbn-mag --num_g
u
ps 1
python3 hetero_rgcn.py --dataset ogbn-mag --num_gp
u
s 1
```
### Resource usage and time cost
...
...
@@ -19,8 +19,8 @@ Below results are roughly collected from an AWS EC2 **g4dn.metal**, 384GB RAM, 9
| Dataset Size | CPU RAM Usage | Num of GPUs | GPU RAM Usage | Time Per Epoch(Training) | Time Per Epoch(Inference: train/val/test set) |
| ------------ | ------------- | ----------- | ---------- | --------- | --------------------------- |
| ~1.1GB | ~4.5GB | 0 | 0GB | ~
4
min
14
s(615it, 2.
41
it/s) | ~0min
2
8s(154it,
5.46
it/s) + ~0min
2
s(16it,
5.48
it/s) + ~0min
2
s(11it,
5.44
it/s) |
| ~1.1GB | ~2GB | 1 | 4.4GB | ~
1min1
5s(615it,
8.11
it/s) | ~0min
27
s(154it,
5.63
it/s) + ~0min
2
s(16it,
5.90
it/s) + ~0min1s(11it,
5
.82it/s) |
| ~1.1GB | ~4.5GB | 0 | 0GB | ~
3
min
55
s(615it, 2.
62
it/s) | ~0min
1
8s(154it,
8.37
it/s) + ~0min
1
s(16it,
8.37
it/s) + ~0min
1
s(11it,
8.88
it/s) |
| ~1.1GB | ~2GB | 1 | 4.4GB | ~5
8
s(615it,
10.43
it/s) | ~0min
18
s(154it,
8.29
it/s) + ~0min
1
s(16it,
8.46
it/s) + ~0min1s(11it,
8
.82it/s) |
### Accuracies
```
...
...
@@ -41,7 +41,7 @@ python3 hetero_rgcn.py --dataset ogb-lsc-mag240m
### Sample on CPU and train/infer on GPU
```
python3 hetero_rgcn.py --dataset ogb-lsc-mag240m --num_g
u
ps 1
python3 hetero_rgcn.py --dataset ogb-lsc-mag240m --num_gp
u
s 1
```
### Resource usage and time cost
...
...
@@ -49,7 +49,7 @@ Below results are roughly collected from an AWS EC2 **g4dn.metal**, 384GB RAM, 9
| Dataset Size | CPU RAM Usage | Num of GPUs | GPU RAM Usage | Time Per Epoch(Training) | Time Per Epoch(Inference: train/val/test set) |
| ------------ | ------------- | ----------- | ---------- | --------- | --------------------------- |
| ~404GB | ~55GB | 0 | 0GB | ~3min5
1
s(1087it,
4.70
it/s) | ~2min2
1
s(272it, 1.
93
it/s) + ~0min2
2
s(34it, 1.
48
it/s) + ~0min1
4
s(22it, 1.
51
it/s) |
| ~404GB | ~55GB | 0 | 0GB | ~3min
2
5s(1087it,
5.29
it/s) | ~2min2
6
s(272it, 1.
86
it/s) + ~0min2
0
s(34it, 1.
62
it/s) + ~0min1
3
s(22it, 1.
68
it/s) |
| ~404GB | ~55GB | 1 | 7GB | ~2min41s(1087it, 6.73it/s) | ~1min52s(272it, 2.41it/s) + ~0min17s(34it, 1.93it/s) + ~0min11s(22it, 1.99it/s) |
### Accuracies
...
...
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