Benchmark intel xpu

[sunjiweiswift]

Added llama for Intel support. The command is still sh benchmark.sh

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/1259

• 📄 Preview Python docs built from this PR

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

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

[facebook-github-bot]

Hi @sunjiweiswift!

Thank you for your pull request and welcome to our community.

Action Required

In order to merge any pull request (code, docs, etc.), we require contributors to sign our Contributor License Agreement, and we don't seem to have one on file for you.

Process

In order for us to review and merge your suggested changes, please sign at https://code.facebook.com/cla. If you are contributing on behalf of someone else (eg your employer), the individual CLA may not be sufficient and your employer may need to sign the corporate CLA.

Once the CLA is signed, our tooling will perform checks and validations. Afterwards, the pull request will be tagged with CLA signed. The tagging process may take up to 1 hour after signing. Please give it that time before contacting us about it.

If you have received this in error or have any questions, please contact us at [email protected]. Thanks!

[jerryzh168]

can you add some test command in summary to show how to run benchmarks for xpu

[sunjiweiswift]

can you add some test command in summary to show how to run benchmarks for xpu

I modified the default_device variable ("default_device = 'cuda' if torch.cuda.is_available() else 'xpu' if torch.xpu.is_available() else 'cpu'") so it will be set to cuda on nvdia gpu and xpu on intel gpu. So I still use the “sh benchmark.sh” command .

[jerryzh168]

@sunjiweiswift are you able to publish some benchmarking results as well? if so please add a new entry similar to https://github.com/pytorch/ao/tree/main/torchao/quantization#uintx-quantization with title "xpu backend"

[sunjiweiswift]

@sunjiweiswift are you able to publish some benchmarking results as well? if so please add a new entry similar to https://github.com/pytorch/ao/tree/main/torchao/quantization#uintx-quantization with title "xpu backend"

Currently these quantizations are not fully supported. I updated the benchmark data of bf16 int8dq int8wo, and wikitext-perplexity has not been tested yet(I encountered a problem of insufficient memory on Intel GPU), and will be updated in subsequent commits
https://github.com/pytorch/ao/pull/1259/files#diff-c15f5d9d386c1899418529df1c6d7622f5f9dd151f193dbca841e40078b6c8bd

[jerryzh168]

please rebase to resolve conflict before landing

[sunjiweiswift]

please rebase to resolve conflict before landing

The conflict has been resolved. Please complete the merge