exponential_ few fixes (1) lambda > 0 (2) mkl kernel to continuous (3) better error log on dtype

[min-jean-cho]

Exponential distribution is continuous. Fixes CPU MKL exponential implementation to exclude integer dtypes.

import torch
dtypes = [torch.uint8, torch.int8, torch.int16, torch.int32, torch.int64]

for dtype in dtypes:
    x = torch.empty(10000, dtype=dtype).exponential_() # should fail !
    print("dtype: ", x.dtype, "sum: ", x.sum())

Additional Context

Related to #92709. This issue propagates to OpInfo of exponential.

AssertionError: The supported dtypes for exponential on device type cpu are incorrect!
The following dtypes worked in forward but are not listed by the OpInfo: {torch.int64, torch.uint8, torch.int8, torch.int16, torch.int32}.

cc @jgong5 @mingfeima @XiaobingSuper @sanchitintel @ashokei @jingxu10

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/92891

• 📄 Preview Python docs built from this PR
• 📄 Preview C++ docs built from this PR
• ❓ Need help or want to give feedback on the CI? Visit the bot commands wiki or our office hours

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

⏳ No Failures, 5 Pending

As of commit a30aa39:
💚 Looks good so far! There are no failures yet. 💚

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

[lezcano]

Could you also add a TORCH_CHECK throwing a more informative error?

[min-jean-cho]

Thanks @lezcano, should we add such check for other exponential distribution implementations as well? Currently, the error is exponential_{device} not implemented for {dtype} for the default cpu, cuda as well as cpu mkl implementations.

[lezcano]

Yeah, that'd be great! If this was implemented properly, all the checks would be shared by all the implementations and then we would dispatch to the actual backend, but I guess this is not the case.

[min-jean-cho]

Thanks @lezcano, I've (1) added better error log on dtype (2) fixed lambda >= 0 to lambda > 0 as discussed in #93053. Please have a look, thanks!

[min-jean-cho]

@pytorchbot rebase -b master

[pytorchmergebot]

@pytorchbot successfully started a rebase job. Check the current status here

[pytorchmergebot]

Successfully rebased minjean/mkl_exponential_continuous onto refs/remotes/origin/master, please pull locally before adding more changes (for example, via git checkout minjean/mkl_exponential_continuous && git pull --rebase)

[min-jean-cho]

@pytorchbot merge

[pytorchmergebot]

Merge started

Your change will be merged once all checks pass (ETA 0-4 Hours).

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Advanced Debugging

Check the merge workflow status
here