Move seed from utils to training

[RdoubleA]

Context

What is the purpose of this PR? Is it to

• add a new feature
• fix a bug
• update tests and/or documentation
• other (please add here)

Addresses #1441.

Changelog

What are the changes made in this PR?

Test plan

Please make sure to do each of the following if applicable to your PR. (If you're not sure about any one of these just ask and we will happily help. We also have a contributing page for some guidance on contributing.)

• run pre-commit hooks and linters (make sure you've first installed via pre-commit install)
• add unit tests for any new functionality
• update docstrings for any new or updated methods or classes
• run unit tests via pytest tests
• run recipe tests via pytest tests -m integration_test
• manually run any new or modified recipes with sufficient proof of correctness
• include relevant commands and any other artifacts in this summary (pastes of loss curves, eval results, etc.)

UX

If your function changed a public API, please add a dummy example of what the user experience will look like when calling it.
Example of docstring: https://github.com/pytorch/torchtune/blob/6a7951f1cdd0b56a9746ef5935106989415f50e3/torchtune/modules/vision_transformer.py#L285
Example in our docs: https://pytorch.org/torchtune/main/tutorials/qat_finetune.html#applying-qat-to-llama3-models

• I did not change any public API;
• I have added an example to docs or docstrings;

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/torchtune/1460

• 📄 Preview Python docs built from this PR

Note: Links to docs will display an error until the docs builds have been completed.

✅ No Failures

As of commit 5a1e4f0 with merge base 069bc4d ():
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 67.56757% with 12 lines in your changes missing coverage. Please review.

Project coverage is 72.02%. Comparing base (0bdd308) to head (1254514).

Files with missing lines	Patch %	Lines
recipes/ppo_full_finetune_single_device.py	0.00%	2 Missing ⚠️
recipes/eleuther_eval.py	0.00%	1 Missing ⚠️
recipes/full_finetune_distributed.py	0.00%	1 Missing ⚠️
recipes/full_finetune_single_device.py	0.00%	1 Missing ⚠️
recipes/generate.py	0.00%	1 Missing ⚠️
recipes/lora_dpo_distributed.py	0.00%	1 Missing ⚠️
recipes/lora_dpo_single_device.py	0.00%	1 Missing ⚠️
recipes/lora_finetune_distributed.py	0.00%	1 Missing ⚠️
recipes/lora_finetune_single_device.py	0.00%	1 Missing ⚠️
recipes/qat_distributed.py	0.00%	1 Missing ⚠️
... and 1 more

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1460      +/-   ##
==========================================
+ Coverage   69.80%   72.02%   +2.22%     
==========================================
  Files         272      272              
  Lines       13053    13053              
==========================================
+ Hits         9111     9401     +290     
+ Misses       3942     3652     -290     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[joecummings]

This may be one of those rare cases in which it belongs actually in a common utils, rather than training.

My reasoning here is that we need this set_seed for generation and eval, as well, which are not training. Therefore, it would be confusing to import from trainiing.

[RdoubleA]

My reasoning here is that we need this set_seed for generation and eval, as well, which are not training. Therefore, it would be confusing to import from trainiing.

A few of the utils APIs fall under this category, like set_device, set_dtype, etc. But I don't have strong opinions either way, if we leave set_seed in utils then I can close this PR.

[joecummings]

My reasoning here is that we need this set_seed for generation and eval, as well, which are not training. Therefore, it would be confusing to import from trainiing.

A few of the utils APIs fall under this category, like set_device, set_dtype, etc. But I don't have strong opinions either way, if we leave set_seed in utils then I can close this PR.

Yeah I'd prefer to leave those in utils or rename to common or something. not sure.

[SalmanMohammadi]

There's torchune.modules.common_utils.py

[joecummings]

There's torchune.modules.common_utils.py

Nah, I don't think these should live nested in modules/. Let's keep it where it is for now and we can discuss if we need a more descriptive name than utils.

[joecummings]

not happy about it, but unblocking for now :/