Utility function for numerical correctness of edge dialect graphs and reference implementations #14036
Conversation
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
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/14036
Note: Links to docs will display an error until the docs builds have been completed. ❌ 159 New Failures, 1 Pending, 45 Unrelated FailuresAs of commit 2744287 with merge base 29cec35 ( NEW FAILURES - The following jobs have failed:
FLAKY - The following jobs failed but were likely due to flakiness present on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
meta-cla
bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D81843001 |
This PR needs a
|
DrJessop
force-pushed
the
export-D81843001
branch
from
f52e06e to
2b48e89
Compare
DrJessop
pushed a commit
to DrJessop/executorch
that referenced
this pull request
Sep 6, 2025
… reference implementations (pytorch#14036) Summary: Created two utility functions 1. Converts an edge dialect graph into one where custom cadence op nodes are replaced with python references 2. Validates the outputs (and optionally intermediates) of the graphs Updated two tests in test_replace_ops_passes to utilize these utility functions. Differential Revision: D81843001
|
This pull request was exported from Phabricator. Differential Revision: D81843001 |
DrJessop
pushed a commit
to DrJessop/executorch
that referenced
this pull request
Sep 6, 2025
… reference implementations (pytorch#14036) Summary: Pull Request resolved: pytorch#14036 Created two utility functions 1. Converts an edge dialect graph into one where custom cadence op nodes are replaced with python references 2. Validates the outputs (and optionally intermediates) of the graphs Updated two tests in test_replace_ops_passes to utilize these utility functions. Differential Revision: D81843001
DrJessop
force-pushed
the
export-D81843001
branch
2 times, most recently
from
a028fc2 to
6b1517d
Compare
DrJessop
pushed a commit
to DrJessop/executorch
that referenced
this pull request
Sep 8, 2025
… reference implementations (pytorch#14036) Summary: Created two utility functions 1. Converts an edge dialect graph into one where custom cadence op nodes are replaced with python references 2. Validates the outputs (and optionally intermediates) of the graphs Updated two tests in test_replace_ops_passes to utilize these utility functions. Differential Revision: D81843001
|
This pull request was exported from Phabricator. Differential Revision: D81843001 |
DrJessop
pushed a commit
to DrJessop/executorch
that referenced
this pull request
Sep 8, 2025
… reference implementations (pytorch#14036) Summary: Pull Request resolved: pytorch#14036 Created two utility functions 1. Converts an edge dialect graph into one where custom cadence op nodes are replaced with python references 2. Validates the outputs (and optionally intermediates) of the graphs Updated two tests in test_replace_ops_passes to utilize these utility functions. Differential Revision: D81843001
DrJessop
force-pushed
the
export-D81843001
branch
from
6b1517d to
1bca92a
Compare
DrJessop
pushed a commit
to DrJessop/executorch
that referenced
this pull request
Sep 8, 2025
… reference implementations (pytorch#14036) Summary: Created two utility functions 1. Converts an edge dialect graph into one where custom cadence op nodes are replaced with python references 2. Validates the outputs (and optionally intermediates) of the graphs Updated two tests in test_replace_ops_passes to utilize these utility functions. Differential Revision: D81843001
DrJessop
pushed a commit
to DrJessop/executorch
that referenced
this pull request
Sep 8, 2025
… reference implementations (pytorch#14036) Summary: Created two utility functions 1. Converts an edge dialect graph into one where custom cadence op nodes are replaced with python references 2. Validates the outputs (and optionally intermediates) of the graphs Updated two tests in test_replace_ops_passes to utilize these utility functions. Differential Revision: D81843001
DrJessop
pushed a commit
to DrJessop/executorch
that referenced
this pull request
Sep 8, 2025
… reference implementations (pytorch#14036) Summary: Created two utility functions 1. Converts an edge dialect graph into one where custom cadence op nodes are replaced with python references 2. Validates the outputs (and optionally intermediates) of the graphs Updated two tests in test_replace_ops_passes to utilize these utility functions. Differential Revision: D81843001
DrJessop
pushed a commit
to DrJessop/executorch
that referenced
this pull request
Sep 8, 2025
… reference implementations (pytorch#14036) Summary: Created two utility functions 1. Converts an edge dialect graph into one where custom cadence op nodes are replaced with python references 2. Validates the outputs (and optionally intermediates) of the graphs Updated two tests in test_replace_ops_passes to utilize these utility functions. Differential Revision: D81843001
DrJessop
pushed a commit
to DrJessop/executorch
that referenced
this pull request
Sep 8, 2025
… reference implementations (pytorch#14036) Summary: Created two utility functions 1. Converts an edge dialect graph into one where custom cadence op nodes are replaced with python references 2. Validates the outputs (and optionally intermediates) of the graphs Updated two tests in test_replace_ops_passes to utilize these utility functions. Differential Revision: D81843001
DrJessop
pushed a commit
to DrJessop/executorch
that referenced
this pull request
Sep 8, 2025
… reference implementations (pytorch#14036) Summary: Created two utility functions 1. Converts an edge dialect graph into one where custom cadence op nodes are replaced with python references 2. Validates the outputs (and optionally intermediates) of the graphs Updated two tests in test_replace_ops_passes to utilize these utility functions. Differential Revision: D81843001
added 9 commits
September 9, 2025 09:25
Summary: Continued support of adding custom Cadence python references Differential Revision: D81720359
Summary: Fixes mismatches between op registration names and implementation names, fixes some type issues in tests where unexpected types are passed in given the op definition. Also fixes an incorrect layernorm meta op (normalized_shape should be list, not int). Tests corrected as well. Tests now use the torch cadence custom op library. Differential Revision: D81738196
…ls_last Summary: The default overload of custom channels last assumes that inputs and weights are permuted and contiguous in memory. Differential Revision: D81842686
Summary: Continued support for reference implementations of all custom Cadence ops. Differential Revision: D81940978
Summary: Quantized fully connected are just aliases for quantized_linear, so created all aliases. Differential Revision: D81942767
Summary: Create a generic quantized relu and decorators for all custom quantized relu ops. Differential Revision: D81948125
Summary: As discussed offline, we don't need a non-per-tensor variant of quantized_add, so removing from ref implementations. Differential Revision: D81950579
Summary: Add type specialized variants of quantized_add_per_tensor Differential Revision: D81951110
Summary: Pull Request resolved: pytorch#14095 Built on top of quantized_linear infrastructure. Differential Revision: D81973532
DrJessop
force-pushed
the
export-D81843001
branch
from
1bca92a to
97fbb37
Compare
DrJessop
pushed a commit
to DrJessop/executorch
that referenced
this pull request
Sep 9, 2025
… reference implementations (pytorch#14036) Summary: Created two utility functions 1. Converts an edge dialect graph into one where custom cadence op nodes are replaced with python references 2. Validates the outputs (and optionally intermediates) of the graphs Updated two tests in test_replace_ops_passes to utilize these utility functions. Differential Revision: D81843001
… reference implementations (pytorch#14036) Summary: Pull Request resolved: pytorch#14036 Created two utility functions 1. Converts an edge dialect graph into one where custom cadence op nodes are replaced with python references 2. Validates the outputs (and optionally intermediates) of the graphs Updated two tests in test_replace_ops_passes to utilize these utility functions. Differential Revision: D81843001
|
This pull request was exported from Phabricator. Differential Revision: D81843001 |
DrJessop
force-pushed
the
export-D81843001
branch
from
97fbb37 to
2744287
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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.
Summary:
Created two utility functions
Updated two tests in test_replace_ops_passes to utilize these utility functions.
Differential Revision: D81843001