Properly import LooseVersion #996
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Summary: This fixes regression introduced by pytorch#57040 Somehow importing `distutils` from `setuptool` caused import of `distutils.versions`, which is not a documented dependency and got change with the release of [setuptools-59.6.0](https://github.com/pypa/setuptools/tree/v59.6.0) We should not rely on that, as `import distutils` never re-imports `distutils.version`, which one can see by observing https://github.com/python/cpython/blob/3.9/Lib/distutils/__init__.py or by running: ``` % python3 -c "import distutils;print(distutils.__version__, dir(distutils))" 3.7.5 ['__builtins__', '__cached__', '__doc__', '__file__', '__loader__', '__name__', '__package__', '__path__', '__spec__', '__version__', 'sys'] % python3 -c "from setuptools import distutils;print(distutils.__version__, dir(distutils))" 3.7.5 ['__builtins__', '__cached__', '__doc__', '__file__', '__loader__', '__name__', '__package__', '__path__', '__spec__', '__version__', 'archive_util', 'ccompiler', 'cmd', 'config', 'core', 'debug', 'dep_util', 'dir_util', 'dist', 'errors', 'extension', 'fancy_getopt', 'file_util', 'filelist', 'log', 'spawn', 'sys', 'sysconfig', 'util', 'version'] ``` Pull Request resolved: pytorch#69904 Reviewed By: albanD, atalman, janeyx99 Differential Revision: D33094453 Pulled By: malfet fbshipit-source-id: aaf1adb7c6f293c4e376ccff21c64cd6ba625e97
pruthvistony
pushed a commit
that referenced
this pull request
Apr 21, 2022
Properly import LooseVersion (pytorch#69904) Summary: This fixes regression introduced by pytorch#57040 Somehow importing `distutils` from `setuptool` caused import of `distutils.versions`, which is not a documented dependency and got change with the release of [setuptools-59.6.0](https://github.com/pypa/setuptools/tree/v59.6.0) We should not rely on that, as `import distutils` never re-imports `distutils.version`, which one can see by observing https://github.com/python/cpython/blob/3.9/Lib/distutils/__init__.py or by running: ``` % python3 -c "import distutils;print(distutils.__version__, dir(distutils))" 3.7.5 ['__builtins__', '__cached__', '__doc__', '__file__', '__loader__', '__name__', '__package__', '__path__', '__spec__', '__version__', 'sys'] % python3 -c "from setuptools import distutils;print(distutils.__version__, dir(distutils))" 3.7.5 ['__builtins__', '__cached__', '__doc__', '__file__', '__loader__', '__name__', '__package__', '__path__', '__spec__', '__version__', 'archive_util', 'ccompiler', 'cmd', 'config', 'core', 'debug', 'dep_util', 'dir_util', 'dist', 'errors', 'extension', 'fancy_getopt', 'file_util', 'filelist', 'log', 'spawn', 'sys', 'sysconfig', 'util', 'version'] ``` Pull Request resolved: pytorch#69904 Reviewed By: albanD, atalman, janeyx99 Differential Revision: D33094453 Pulled By: malfet fbshipit-source-id: aaf1adb7c6f293c4e376ccff21c64cd6ba625e97
jithunnair-amd
pushed a commit
to jithunnair-amd/pytorch
that referenced
this pull request
Sep 20, 2022
Properly import LooseVersion (pytorch#69904) Summary: This fixes regression introduced by pytorch#57040 Somehow importing `distutils` from `setuptool` caused import of `distutils.versions`, which is not a documented dependency and got change with the release of [setuptools-59.6.0](https://github.com/pypa/setuptools/tree/v59.6.0) We should not rely on that, as `import distutils` never re-imports `distutils.version`, which one can see by observing https://github.com/python/cpython/blob/3.9/Lib/distutils/__init__.py or by running: ``` % python3 -c "import distutils;print(distutils.__version__, dir(distutils))" 3.7.5 ['__builtins__', '__cached__', '__doc__', '__file__', '__loader__', '__name__', '__package__', '__path__', '__spec__', '__version__', 'sys'] % python3 -c "from setuptools import distutils;print(distutils.__version__, dir(distutils))" 3.7.5 ['__builtins__', '__cached__', '__doc__', '__file__', '__loader__', '__name__', '__package__', '__path__', '__spec__', '__version__', 'archive_util', 'ccompiler', 'cmd', 'config', 'core', 'debug', 'dep_util', 'dir_util', 'dist', 'errors', 'extension', 'fancy_getopt', 'file_util', 'filelist', 'log', 'spawn', 'sys', 'sysconfig', 'util', 'version'] ``` Pull Request resolved: pytorch#69904 Reviewed By: albanD, atalman, janeyx99 Differential Revision: D33094453 Pulled By: malfet fbshipit-source-id: aaf1adb7c6f293c4e376ccff21c64cd6ba625e97
jithunnair-amd
pushed a commit
that referenced
this pull request
Sep 28, 2022
Properly import LooseVersion (pytorch#69904) Summary: This fixes regression introduced by pytorch#57040 Somehow importing `distutils` from `setuptool` caused import of `distutils.versions`, which is not a documented dependency and got change with the release of [setuptools-59.6.0](https://github.com/pypa/setuptools/tree/v59.6.0) We should not rely on that, as `import distutils` never re-imports `distutils.version`, which one can see by observing https://github.com/python/cpython/blob/3.9/Lib/distutils/__init__.py or by running: ``` % python3 -c "import distutils;print(distutils.__version__, dir(distutils))" 3.7.5 ['__builtins__', '__cached__', '__doc__', '__file__', '__loader__', '__name__', '__package__', '__path__', '__spec__', '__version__', 'sys'] % python3 -c "from setuptools import distutils;print(distutils.__version__, dir(distutils))" 3.7.5 ['__builtins__', '__cached__', '__doc__', '__file__', '__loader__', '__name__', '__package__', '__path__', '__spec__', '__version__', 'archive_util', 'ccompiler', 'cmd', 'config', 'core', 'debug', 'dep_util', 'dir_util', 'dist', 'errors', 'extension', 'fancy_getopt', 'file_util', 'filelist', 'log', 'spawn', 'sys', 'sysconfig', 'util', 'version'] ``` Pull Request resolved: pytorch#69904 Reviewed By: albanD, atalman, janeyx99 Differential Revision: D33094453 Pulled By: malfet fbshipit-source-id: aaf1adb7c6f293c4e376ccff21c64cd6ba625e97
akashveramd
pushed a commit
that referenced
this pull request
Jun 13, 2025
Noticed that in the memory snapshot when running the Llama3-8B models, the cross entropy loss is causing huge spikes in peak memory: <img width="474" alt="image" src="https://github.com/user-attachments/assets/31ea0f9a-7734-4454-90e5-e3a4ed0b5db3" /> Replace the cross entropy loss function with a chunked version can help reduce peak memory by 7GiB without affecting training throughput. This is inspired by the torchtune [implementation](https://github.com/pytorch/torchtune/blob/c3703482bde72e572b535d3f7c43c81e94164ebc/torchtune/modules/loss/ce_chunked_output_loss.py#L13). **Baseline run** ``` ... [rank0]:[titan] 2025-03-20 08:19:11,539 - root - INFO - step: 30 loss: 7.7158 memory: 49.58GiB(52.20%) tps: 5,966 tflops: 345.53 mfu: 34.94% [rank0]:[titan] 2025-03-20 08:19:25,291 - root - INFO - step: 40 loss: 7.3677 memory: 49.58GiB(52.20%) tps: 5,957 tflops: 345.01 mfu: 34.88% [rank0]:[titan] 2025-03-20 08:19:39,040 - root - INFO - step: 50 loss: 7.0890 memory: 49.58GiB(52.20%) tps: 5,959 tflops: 345.10 mfu: 34.89% [rank0]:[titan] 2025-03-20 08:19:52,794 - root - INFO - step: 60 loss: 6.8336 memory: 49.58GiB(52.20%) tps: 5,957 tflops: 344.99 mfu: 34.88% ... ``` **With the changes** ``` ... [rank0]:[titan] 2025-03-20 08:29:30,780 - root - INFO - step: 30 loss: 7.7423 memory: 42.26GiB(44.49%) tps: 5,957 tflops: 345.01 mfu: 34.89% [rank0]:[titan] 2025-03-20 08:29:44,554 - root - INFO - step: 40 loss: 7.3368 memory: 42.26GiB(44.49%) tps: 5,948 tflops: 344.49 mfu: 34.83% [rank0]:[titan] 2025-03-20 08:29:58,318 - root - INFO - step: 50 loss: 7.0574 memory: 42.26GiB(44.49%) tps: 5,953 tflops: 344.74 mfu: 34.86% [rank0]:[titan] 2025-03-20 08:30:12,080 - root - INFO - step: 60 loss: 6.8472 memory: 42.26GiB(44.49%) tps: 5,953 tflops: 344.79 mfu: 34.86% ... ``` **cc**: @lessw2020 @weifengpy @wconstab @yf225
akashveramd
pushed a commit
that referenced
this pull request
Jun 13, 2025
To enable chunked loss, I refactor the `train_spec.loss_fn` to `train_spec.build_loss_fn` as suggested in #996 Also: - Unified chunked loss implementation with the `cross_entropy_loss` name, since default is a special case with chunk=1. Identical loss and memory: <img width="1257" alt="Screenshot 2025-03-30 at 13 03 44" src="https://github.com/user-attachments/assets/11c07849-2340-4b42-ad5c-d96657ee53b9" /> - Enable compiling the loss function as it's working fine for me, and result in reduction of memory and slightly better througput. <img width="1275" alt="Screenshot 2025-03-30 at 16 47 00" src="https://github.com/user-attachments/assets/09edaf16-1770-48da-87b2-e4de29e7519c" /> While we only have one type of loss, I create a field for `loss.name` with default to 'cross_entropy' to be fork-friendly anyway. Co-authored-by: tianyu-l <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.