Uh oh!
There was an error while loading. Please reload this page.
Updating documentation for cmake dtype selective build - #12112
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/12112
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit e9745f4 with merge base adb5318 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
BujSet
commented
Jun 30, 2025
@pytorchbot label "release notes: none" |
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
BujSet
commented
Jun 30, 2025
cc @psiddh |
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
| ### Select ops from model | ||
| This API lets users pass in a pte file of an exported model. When used, the pte file will be parsed to generate a yaml file that enumerates the operators and dtypes used in the model. |
There was a problem hiding this comment.
Is the fact that a yaml file is generated relevant to the end user? The above documentation makes it seem like the model yaml is not a user facing construct, only the pte, or the root ops list are exposed methods of passing selective information.
Also in general I would say "root" ops is misleading since ET doesnt have transitive ops. We should just call it the op list.
There was a problem hiding this comment.
When using the model API, the generated YAML can be as an input to the selective build process, like when using the yaml API directly. The difference being that a user doesn't have to manually craft a yaml, but can pass in a pte.
I agree, root ops is not a great name, but it seems to be used in other places as well, e.g. https://github.com/BujSet/executorch/blob/main/codegen/tools/gen_oplist.py#L234-L243. For consistency, I think we should keep it as root ops, and a separate PR should refactor this to something more meaningful, what do you think?
Uh oh!
There was an error while loading. Please reload this page.
a6f2443 to
8da8a29Compare8da8a29 to
e9745f4Comparemergennachin
commented
Jun 30, 2025
@lucylq@BujSet@larryliu0820 should this part of the cherry-pick for 0.7? |
lucylq
commented
Jun 30, 2025
Yes, I think we can include it as all the code changes are there. @BujSet could you cherry-pick it following the instructions in #11075 ? |
BujSet
commented
Jun 30, 2025
@pytorchbot cherry-pick --onto release/0.7 -c docs |
### Summary Updating documentation for cmake dtype selective build with model API. Table at `selective_build_model_doc/docs/source/kernel-library-selective-build.md` now looks like:  Fixes#12065 cc @mergennachin@byjlw (cherry picked from commit f673a4b)
pytorchbot
commented
Jun 30, 2025
Cherry picking #12112The cherry pick PR is at #12123 The following tracker issues are updated: Details for Dev Infra teamRaised by workflow job |
Updating documentation for cmake dtype selective build with model API. Table at `selective_build_model_doc/docs/source/kernel-library-selective-build.md` now looks like:  Fixespytorch#12065 cc @mergennachin@byjlw
Summary
Updating documentation for cmake dtype selective build with model API. Table at
selective_build_model_doc/docs/source/kernel-library-selective-build.mdnow looks like:Fixes#12065
cc @mergennachin@byjlw