Checkpoint Function Not Well Tested For Old Pytorch Versions #7265
Comments
KumoLiu
added a commit
to KumoLiu/MONAI
that referenced
this issue
Nov 29, 2023
Signed-off-by: KumoLiu <yunl@nvidia.com>
7 tasks
|
A simple solution would be to skip the test before 1.10. What do you think? |
ericspod
pushed a commit
that referenced
this issue
Nov 30, 2023
Fixes #7265. ### Types of changes <!--- Put an `x` in all the boxes that apply, and remove the not applicable items --> - [x] 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. --------- Signed-off-by: KumoLiu <yunl@nvidia.com>
marksgraham
pushed a commit
to marksgraham/MONAI
that referenced
this issue
Jan 30, 2024
Fixes Project-MONAI#7265. ### Types of changes <!--- Put an `x` in all the boxes that apply, and remove the not applicable items --> - [x] 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. --------- Signed-off-by: KumoLiu <yunl@nvidia.com> Signed-off-by: Mark Graham <markgraham539@gmail.com>
juampatronics
pushed a commit
to juampatronics/MONAI
that referenced
this issue
Mar 25, 2024
Fixes Project-MONAI#7265. ### Types of changes <!--- Put an `x` in all the boxes that apply, and remove the not applicable items --> - [x] 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. --------- Signed-off-by: KumoLiu <yunl@nvidia.com> Signed-off-by: Juan Pablo de la Cruz Gutiérrez <juampatronics@gmail.com>
Yu0610
pushed a commit
to Yu0610/MONAI
that referenced
this issue
Apr 11, 2024
Fixes Project-MONAI#7265. ### Types of changes <!--- Put an `x` in all the boxes that apply, and remove the not applicable items --> - [x] 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. --------- Signed-off-by: KumoLiu <yunl@nvidia.com> Signed-off-by: Yu0610 <612410030@alum.ccu.edu.tw>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Describe the bug
The function
torch.utils.checkpoint.checkpointdoesn't have ause_reentrantargument in Pytorch<=1.10. This function is used with this argument inSwinUNETRbut tests don't cover the case where using checkpointing is enabled. The tests should be enhanced to account for this, or we drop support for older versions of Pytorch. This is related to an issue in PR #7260.To Reproduce
Steps to reproduce the behavior:
SwinUNETRwithuse_checkpoint=Trueand attempt a forward pass using an older version of Pytorch.Expected behavior
Forward pass should work as normal.
Environment
Pytorch<=1.10.
The text was updated successfully, but these errors were encountered: