Skip to content

4814 Update PyTorch base docker to 22.07#4815

Merged
wyli merged 13 commits intoProject-MONAI:devfrom
Nic-Ma:4814-update-pytorch-docker
Aug 3, 2022
Merged

4814 Update PyTorch base docker to 22.07#4815
wyli merged 13 commits intoProject-MONAI:devfrom
Nic-Ma:4814-update-pytorch-docker

Conversation

@Nic-Ma
Copy link
Copy Markdown
Contributor

@Nic-Ma Nic-Ma commented Aug 2, 2022

Fixes #4814 .

Description

This PR updated the PyTorch base docker to 22.07.

Status

Ready

Types of changes

  • Non-breaking change (fix or new feature that would not break existing functionality).
  • Breaking change (fix or new feature that would cause existing functionality to change).
  • New tests added to cover the changes.
  • Integration tests passed locally by running ./runtests.sh -f -u --net --coverage.
  • Quick tests passed locally by running ./runtests.sh --quick --unittests --disttests.
  • In-line docstrings updated.
  • Documentation updated, tested make html command in the docs/ folder.

@Nic-Ma
Copy link
Copy Markdown
Contributor Author

Nic-Ma commented Aug 2, 2022

/integration-test

@Nic-Ma Nic-Ma requested review from YanxuanLiu and wyli August 2, 2022 07:59
@Nic-Ma
Copy link
Copy Markdown
Contributor Author

Nic-Ma commented Aug 2, 2022

Hi @YanxuanLiu ,

Could you please help update the corresponding Blossom CI config?

Thanks in advance.

@YanxuanLiu
Copy link
Copy Markdown
Collaborator

/build

@Nic-Ma
Copy link
Copy Markdown
Contributor Author

Nic-Ma commented Aug 2, 2022

Hi @Can-Zhao @wyli ,

I saw an issue in the box utils test:

======================================================================
ERROR: test_value_35 (__main__.TestCreateBoxList)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/opt/conda/lib/python3.8/site-packages/parameterized/parameterized.py", line 533, in standalone_func
    return func(*(a + p.args), **p.kwargs)
  File "tests/test_box_utils.py", line 208, in test_value
    nms_box = non_max_suppression(
  File "/workspace/data/medical/MONAI/monai/data/box_utils.py", line 1082, in non_max_suppression
    idxs = idxs[to_keep_idx]
RuntimeError: indices should be either on cpu or on the same device as the indexed tensor (cpu)

So I slightly changed the box utils function to fix it, could you please help review it?
Maybe it's not correct, please feel free to submit other fixes, and I didn't check deeper why the previous docker can pass the tests.

Thanks in advance.

@Nic-Ma Nic-Ma requested a review from Can-Zhao August 2, 2022 10:19
@Nic-Ma
Copy link
Copy Markdown
Contributor Author

Nic-Ma commented Aug 2, 2022

/build

4 similar comments
@YanxuanLiu
Copy link
Copy Markdown
Collaborator

/build

@Nic-Ma
Copy link
Copy Markdown
Contributor Author

Nic-Ma commented Aug 2, 2022

/build

@YanxuanLiu
Copy link
Copy Markdown
Collaborator

/build

@wyli
Copy link
Copy Markdown
Contributor

wyli commented Aug 2, 2022

/build

@Nic-Ma Nic-Ma marked this pull request as ready for review August 2, 2022 14:14
@Nic-Ma Nic-Ma changed the title [WIP] 4814 Update PyTorch base docker to 22.07 4814 Update PyTorch base docker to 22.07 Aug 2, 2022
@Nic-Ma Nic-Ma requested review from ericspod and rijobro August 2, 2022 14:15
@YanxuanLiu
Copy link
Copy Markdown
Collaborator

/build

@Can-Zhao
Copy link
Copy Markdown
Collaborator

Can-Zhao commented Aug 3, 2022

Hi @Can-Zhao @wyli ,

I saw an issue in the box utils test:

======================================================================
ERROR: test_value_35 (__main__.TestCreateBoxList)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/opt/conda/lib/python3.8/site-packages/parameterized/parameterized.py", line 533, in standalone_func
    return func(*(a + p.args), **p.kwargs)
  File "tests/test_box_utils.py", line 208, in test_value
    nms_box = non_max_suppression(
  File "/workspace/data/medical/MONAI/monai/data/box_utils.py", line 1082, in non_max_suppression
    idxs = idxs[to_keep_idx]
RuntimeError: indices should be either on cpu or on the same device as the indexed tensor (cpu)

So I slightly changed the box utils function to fix it, could you please help review it? Maybe it's not correct, please feel free to submit other fixes, and I didn't check deeper why the previous docker can pass the tests.

Thanks in advance.

Thanks Nic. I pushed another commit

@Nic-Ma
Copy link
Copy Markdown
Contributor Author

Nic-Ma commented Aug 3, 2022

Hi @wyli ,

All the integration tests passed locally on V100 GPU with this PyTorch 22.07 docker.

Thanks.

@Nic-Ma
Copy link
Copy Markdown
Contributor Author

Nic-Ma commented Aug 3, 2022

/black

@Nic-Ma
Copy link
Copy Markdown
Contributor Author

Nic-Ma commented Aug 3, 2022

/build

Copy link
Copy Markdown
Contributor

@wyli wyli left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

looks good to me if the cuda preview version doesn't affect our modules

@Nic-Ma
Copy link
Copy Markdown
Contributor Author

Nic-Ma commented Aug 3, 2022

let's wait a while for @deepib 's confirmation.

Thanks.

@YanxuanLiu
Copy link
Copy Markdown
Collaborator

/build

1 similar comment
@Nic-Ma
Copy link
Copy Markdown
Contributor Author

Nic-Ma commented Aug 3, 2022

/build

@Nic-Ma
Copy link
Copy Markdown
Contributor Author

Nic-Ma commented Aug 3, 2022

Hi @YanxuanLiu ,

I clicked the update button and got the DCO error now, have you changed some CI setting about signing?

Thanks.

@Nic-Ma Nic-Ma force-pushed the 4814-update-pytorch-docker branch from 3d06707 to edf32f4 Compare August 3, 2022 07:59
@pxLi
Copy link
Copy Markdown
Contributor

pxLi commented Aug 3, 2022

from the error page

Summary
Commit sha: 18c5d16, Author: Can Zhao, Committer: GitHub; The sign-off is missing.

@Nic-Ma
Copy link
Copy Markdown
Contributor Author

Nic-Ma commented Aug 3, 2022

/build

1 similar comment
@YanxuanLiu
Copy link
Copy Markdown
Collaborator

/build

@YanxuanLiu
Copy link
Copy Markdown
Collaborator

/build

re-build to run with latest image

@Nic-Ma
Copy link
Copy Markdown
Contributor Author

Nic-Ma commented Aug 3, 2022

/build

@Nic-Ma
Copy link
Copy Markdown
Contributor Author

Nic-Ma commented Aug 3, 2022

Hi @wyli @YanxuanLiu ,

Thanks for the update in the Blossom CI, is it OK to merge this PR now?

Thanks.

@wyli
Copy link
Copy Markdown
Contributor

wyli commented Aug 3, 2022

/build

@wyli wyli enabled auto-merge (squash) August 3, 2022 15:25
@wyli wyli merged commit ef05b24 into Project-MONAI:dev Aug 3, 2022
@YanxuanLiu
Copy link
Copy Markdown
Collaborator

Hi @wyli @YanxuanLiu ,

Thanks for the update in the Blossom CI, is it OK to merge this PR now?

Thanks.

Hi @Nic-Ma ,
Blossom CI modifications will be merged after all tests done.

@YanxuanLiu
Copy link
Copy Markdown
Collaborator

Hi @wyli @YanxuanLiu ,
Thanks for the update in the Blossom CI, is it OK to merge this PR now?
Thanks.

Hi @Nic-Ma , Blossom CI modifications will be merged after all tests done.

blossom ci changes merged

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Update PyTorch base container to 22.07

5 participants