Add some basic xnnpack recipes #10035
Open
Conversation
This file contains 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
Differential Revision: [D72085170](https://our.internmc.facebook.com/intern/diff/D72085170/) [ghstack-poisoned]
tarun292
requested review from
digantdesai,
mcr229 and
kimishpatel
as code owners
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/10035
Note: Links to docs will display an error until the docs builds have been completed. ❌ 62 New Failures, 11 Unrelated FailuresAs of commit 910ac24 with merge base 1facfa9 ( 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. |
facebook-github-bot
added
the
CLA Signed
This was referenced Apr 10, 2025
|
This pull request was exported from Phabricator. Differential Revision: D72085170 |
This PR needs a
|
mcr229
reviewed
Apr 10, 2025
| is_per_channel=True, is_dynamic=True | ||
| ) | ||
| quantizer.set_global(operator_config) | ||
| DuplicateDynamicQuantChainPass |
There was a problem hiding this comment.
i've actually done some work to remove the need for this. I guess it's ok to have for now as it should still work.
There was a problem hiding this comment.
Let's remove it if not needed?
kimishpatel
reviewed
Apr 11, 2025
|
|
||
| ) | ||
|
|
||
| def get_dynamic_quant_recipe() -> ExportRecipe: |
There was a problem hiding this comment.
organizationally maybe quant recipes can be in a separate folder?
digantdesai
approved these changes
Apr 11, 2025
| ) | ||
| from executorch.exir import ExportRecipe | ||
|
|
||
| def get_generic_fp32_cpu_recipe() -> ExportRecipe: |
There was a problem hiding this comment.
if namespaced to XNNPACK then cpu may not be needed?
Suggested change
| def get_generic_fp32_cpu_recipe() -> ExportRecipe: | |
| def get_fp32_recipe() -> ExportRecipe: |
Comment on lines
+45
to
+46
| "FP32_CPU_ACCELERATED_RECIPE": get_generic_fp32_cpu_recipe, | ||
| "DYNAMIC_QUANT_CPU_ACCELERATED_RECIPE": get_dynamic_quant_recipe, |
There was a problem hiding this comment.
nit
Suggested change
| "FP32_CPU_ACCELERATED_RECIPE": get_generic_fp32_cpu_recipe, | |
| "DYNAMIC_QUANT_CPU_ACCELERATED_RECIPE": get_dynamic_quant_recipe, | |
| "FP32_RECIPE": get_fp32_recipe, | |
| "DYNAMIC_QUANT_RECIPE": get_dynamic_quant_recipe, |
| name = "fp32_recipe", | ||
| quantizer = None, | ||
| partitioners=[XnnpackPartitioner()], | ||
|
|
Comment on lines
+20
to
+22
| quantizer = XNNPACKQuantizer() | ||
| operator_config = get_symmetric_quantization_config(is_per_channel=False) | ||
| quantizer.set_global(operator_config) |
There was a problem hiding this comment.
Suggested change
| quantizer = XNNPACKQuantizer() | |
| operator_config = get_symmetric_quantization_config(is_per_channel=False) | |
| quantizer.set_global(operator_config) |
| is_per_channel=True, is_dynamic=True | ||
| ) | ||
| quantizer.set_global(operator_config) | ||
| DuplicateDynamicQuantChainPass |
There was a problem hiding this comment.
Let's remove it if not needed?
| operator_config = get_symmetric_quantization_config(is_per_channel=False) | ||
| quantizer.set_global(operator_config) | ||
| return ExportRecipe( | ||
| name = "fp32_recipe", |
There was a problem hiding this comment.
nit
Suggested change
| name = "fp32_recipe", | |
| name = "fp32", |
Comment on lines
+13
to
+17
| def setUp(self) -> None: | ||
| super().setUp() | ||
|
|
||
| def tearDown(self) -> None: | ||
| super().tearDown() |
| example_inputs=example_inputs, | ||
| export_recipe=get_xnnpack_recipe("FP32_CPU_ACCELERATED_RECIPE") | ||
| ) | ||
| export_session.export() |
There was a problem hiding this comment.
will this raise? should we catch it?
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.
Stack from ghstack (oldest at bottom):
Differential Revision: D72085170