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
ColossalAI
Commits
e13293bb
Commit
e13293bb
authored
Feb 24, 2022
by
アマデウス
Committed by
Frank Lee
Mar 11, 2022
Browse files
fixed CI dataset directory; fixed import error of 2.5d accuracy (#255)
parent
219df6e6
Changes
2
Show whitespace changes
Inline
Side-by-side
Showing
2 changed files
with
3 additions
and
3 deletions
+3
-3
.github/workflows/PR_CI.yml
.github/workflows/PR_CI.yml
+2
-2
colossalai/nn/metric/accuracy_2p5d.py
colossalai/nn/metric/accuracy_2p5d.py
+1
-1
No files found.
.github/workflows/PR_CI.yml
View file @
e13293bb
...
@@ -28,7 +28,7 @@ jobs:
...
@@ -28,7 +28,7 @@ jobs:
runs-on
:
[
self-hosted
,
gpu
]
runs-on
:
[
self-hosted
,
gpu
]
container
:
container
:
image
:
nvcr.io/nvidia/pytorch:21.07-py3
image
:
nvcr.io/nvidia/pytorch:21.07-py3
options
:
--gpus all --rm --ipc=host -v /data/cifar-10:/data/cifar-10
options
:
--gpus all --rm --ipc=host -v /data/
scratch/
cifar-10:/data/
scratch/
cifar-10
timeout-minutes
:
20
timeout-minutes
:
20
steps
:
steps
:
-
name
:
Setup Environment
-
name
:
Setup Environment
...
@@ -48,7 +48,7 @@ jobs:
...
@@ -48,7 +48,7 @@ jobs:
run
:
|
run
:
|
pytest tests
pytest tests
env
:
env
:
DATA
:
/data/cifar-10
DATA
:
/data/
scratch/
cifar-10
format_check
:
format_check
:
name
:
Format Check
name
:
Format Check
...
...
colossalai/nn/metric/accuracy_2p5d.py
View file @
e13293bb
import
torch
import
torch
from
colossalai.nn.layer.parallel_2p5d
import
reduce_by_batch_2p5d
from
colossalai.nn.layer.parallel_2p5d
import
reduce_by_batch_2p5d
,
split_tensor_2p5d
from
torch
import
nn
from
torch
import
nn
from
._utils
import
calc_acc
from
._utils
import
calc_acc
...
...
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