Repository navigation
[executorch][native] Record embedding op samples - #25048
Open
pytorchbot wants to merge 1 commit into
Open
pytorchbot wants to merge 1 commit into
pytorchbot wants to merge 1 commit into
Conversation
Pull Request resolved: #24974 Embedding becomes the second op in the op samples, with a case module like linear's: the backend test suite's embedding models, LLM vocabulary and token shapes, leading batch dims, a scalar index, the first and last rows, int32 indices, `padding_idx`, one table per GGUF super-scale family, and dynamic token and batch dims. It fits the float, int4 weight-only and GGUF schemes. `draws` and `chain` take a `sampler`, standard normal by default, so a case can draw integer indices. `Parallel` moves to `cases.py` for both ops, and its modules are now `branches`, which renames the constants of linear's `scale_families` packages; their values are unchanged. Only weight-only packages with an `nn.Linear` in a model that fits the scheme get `golden-int8_act`, because without one it would repeat `golden`; `Scheme.adds_int8_act` decides this from the modules `int8_activations` quantizes. The native test workflow (`test-native.yml`) and the CMake sample rule now also depend on the backend test suite's `test_embedding.py`, so a change there rebuilds and retests the samples, as `test_linear.py` already does. A new ggml-CPU sample test, `ggml_sample_test_no_linear` in Buck and `native_ggml_sample_test_no_linear` in CTest and the OSS ggml job, judges every op but linear without activation quantization, so CI also covers embedding on plain ggml-CPU. Linear keeps only the act_quant test. Until the next diff, both CPU CI targets exclude the 22 int4 weight-only f32 embedding tests that ggml fails. Authored with Claude Code. Differential Revision: [D130869556](https://our.internmc.facebook.com/intern/diff/D130869556/) **NOTE FOR REVIEWERS**: This PR has internal Meta-specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D130869556/)! ghstack-source-id: 472753701
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/25048
Note: Links to docs will display an error until the docs builds have been completed. ❗ 1 Active SEVsThere are 1 currently active SEVs. If your PR is affected, please view them below: ✅ You can merge normally! (4 Unrelated Failures)As of commit 1e0819a with merge base fcc9575 ( BROKEN TRUNK - The following jobs failed but were present on the merge base:👉 Rebase onto the `viable/strict` branch to avoid these failures
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
meta-cla
Bot
added
the
CLA Signed
This branch was previously deployed
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
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.
This PR was created by the merge bot to help merge the original PR into the main branch.
ghstack PR number: #24974 by @digantdesai
^ Please use this as the source of truth for the PR details, comments, and reviews
ghstack PR base: https://github.com/pytorch/executorch/tree/gh/digantdesai/203/base
ghstack PR head: https://github.com/pytorch/executorch/tree/gh/digantdesai/203/head
Merge bot PR base: https://github.com/pytorch/executorch/tree/gh/digantdesai/202/orig
Merge bot PR head: https://github.com/pytorch/executorch/tree/gh/digantdesai/203/orig
Differential Revision: D130869556
@diff-train-skip-merge