Add RemoveAssert pass to remove _assert_tensor_metadata nodes
#7277
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/7277
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit ebb9eaf with merge base 037a0d3 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D67057219 |
yushangdi
force-pushed
the
export-D67057219
branch
from
80526dd to
a5bcbd2
Compare
yushangdi
added a commit
to yushangdi/executorch
that referenced
this pull request
Dec 12, 2024
…rch#7277) Summary: `_assert_tensor_metadata` nodes is added to the result of exported graphs in D66988295. (More background in T209705957, mostly to relax aten.to constraint). Add a pass to remove this op when calling to_edge. Differential Revision: D67057219
|
This pull request was exported from Phabricator. Differential Revision: D67057219 |
yushangdi
added a commit
to yushangdi/executorch
that referenced
this pull request
Dec 12, 2024
…rch#7277) Summary: `_assert_tensor_metadata` nodes is added to the result of exported graphs in D66988295. (More background in T209705957, mostly to relax aten.to constraint). Add a pass to remove this op when calling to_edge. Differential Revision: D67057219
yushangdi
force-pushed
the
export-D67057219
branch
from
a5bcbd2 to
60db2c8
Compare
|
This pull request was exported from Phabricator. Differential Revision: D67057219 |
yushangdi
force-pushed
the
export-D67057219
branch
from
60db2c8 to
6c9cbbf
Compare
yushangdi
added a commit
to yushangdi/executorch
that referenced
this pull request
Dec 13, 2024
…rch#7277) Summary: `_assert_tensor_metadata` nodes is added to the result of exported graphs in D66988295. (More background in T209705957, mostly to relax aten.to constraint). Add a pass to remove this op when calling to_edge. Differential Revision: D67057219
|
This pull request was exported from Phabricator. Differential Revision: D67057219 |
yushangdi
added a commit
to yushangdi/executorch
that referenced
this pull request
Dec 13, 2024
…rch#7277) Summary: `_assert_tensor_metadata` nodes is added to the result of exported graphs in D66988295. (More background in T209705957, mostly to relax aten.to constraint). Add a pass to remove this op when calling to_edge. Differential Revision: D67057219
yushangdi
force-pushed
the
export-D67057219
branch
from
6c9cbbf to
1fe46a4
Compare
|
This pull request was exported from Phabricator. Differential Revision: D67057219 |
…rch#7277) Summary: `_assert_tensor_metadata` nodes is added to the result of exported graphs in D66988295. (More background in T209705957, mostly to relax aten.to constraint). Add a pass to remove this op when calling to_edge. Differential Revision: D67057219
yushangdi
force-pushed
the
export-D67057219
branch
from
1fe46a4 to
ebb9eaf
Compare
|
This pull request was exported from Phabricator. Differential Revision: D67057219 |
tarun292
approved these changes
Dec 16, 2024
yushangdi
added a commit
to yushangdi/pytorch
that referenced
this pull request
Dec 16, 2024
Summary: if we have a.to(b), and b has a different dtype with a, then it must be a copy. In this case, we do not need to freeze the tensor. Instead, we use torch.ops.aten._assert_tensor_metadata.default to ensure that a must not have the same dtype as b. We need to remove ` torch.ops.aten._assert_tensor_metadata.default` for executorch. Update to pin to include pytorch/executorch#7277. Test Plan: ``` buck2 run 'fbcode//mode/dev-nosan' fbcode//caffe2/test:test_export -- -r test_float_conversion buck2 run 'fbcode//mode/dev-nosan' fbcode//caffe2/test:test_export -- -r test_device_to_mutation_float ADD_NEW_STABILITY_CONFIGS=true buck2 run 'fbcode//mode/opt' fbcode//aps_models/ads/modules/module_stability_tests/ctr_cvr_task_arch_stability_tests:inference_stability_tests buck2 run 'fbcode//mode/dev-nosan' fbcode//deeplearning/aot_inductor/cpu/test:cpu_lowering_utils_test buck2 run 'fbcode//mode/dev-nosan' //executorch/exir/tests:test_memory_format_ops_pass ``` Reviewed By: bdhirsh Differential Revision: D66988295
yushangdi
added a commit
to yushangdi/pytorch
that referenced
this pull request
Dec 18, 2024
Summary: if we have a.to(b), and b has a different dtype with a, then it must be a copy. In this case, we do not need to freeze the tensor. Instead, we use torch.ops.aten._assert_tensor_metadata.default to ensure that a must not have the same dtype as b. We need to remove ` torch.ops.aten._assert_tensor_metadata.default` for executorch. Update to pin to include pytorch/executorch#7277. Test Plan: ``` buck2 run 'fbcode//mode/dev-nosan' fbcode//caffe2/test:test_export -- -r test_float_conversion buck2 run 'fbcode//mode/dev-nosan' fbcode//caffe2/test:test_export -- -r test_device_to_mutation_float ADD_NEW_STABILITY_CONFIGS=true buck2 run 'fbcode//mode/opt' fbcode//aps_models/ads/modules/module_stability_tests/ctr_cvr_task_arch_stability_tests:inference_stability_tests buck2 run 'fbcode//mode/dev-nosan' fbcode//deeplearning/aot_inductor/cpu/test:cpu_lowering_utils_test buck2 run 'fbcode//mode/dev-nosan' //executorch/exir/tests:test_memory_format_ops_pass buck test aps_models/ads/modules/module_stability_tests/histogram_binning_calibration_tests:calibration_by_feature_stability_test -- --env OVERRIDE_IR_CONFIGS=true buck test aps_models/ads/modules/module_stability_tests/histogram_binning_calibration_tests:calibration_stability_test -- test_stability_config --env OVERRIDE_IR_CONFIGS=true buck2 test 'fbcode//mode/opt' fbcode//aps_models/ads/modules/module_stability_tests/soft_max_loss_with_class_weights:stability_tests -- test_stability_config --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/loss_stability_tests:distillation_stability_tests -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/loss_stability_tests:stability_tests - aps_models.ads.modules.module_stability_tests.loss_stability_tests.ctr_cvr_logistic_regression_loss_stability_test.CtrCvrLogisticRegressionLossStabilityTest: test_cider_loss -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/feature_perturbation_stability_tests:sparse_feature_perturbation_test -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/ctr_cvr_task_arch_stability_tests:train_stability_tests -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/ctr_cvr_task_arch_stability_tests:inference_stability_tests -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/ctr_cvr_task_arch_stability_tests:histogram_binning_stability_test -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/aggregator_stability_tests:replicate_aggregator_stability_test -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/aggregator_stability_tests:label_aggregator_stability_test -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/aggregator_stability_tests:dral_aggregator_stability_test -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/adjust_task_weight_stability_tests:task_weight_stability_test -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/loss_modules/module_stability_tests/cider_stability_tests:stability_tests -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/py2pt/calibration/module_stability_tests/histogram_binning_calibration_stability_tests:stability_tests -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/py2pt/calibration/module_stability_tests/histogram_binning_calibration_by_feature_stability_tests:stability_tests -- --env OVERRIDE_IR_CONFIGS=true ``` Reviewed By: bdhirsh, shruthign Differential Revision: D66988295
yushangdi
added a commit
to yushangdi/pytorch
that referenced
this pull request
Dec 18, 2024
Summary: if we have a.to(b), and b has a different dtype with a, then it must be a copy. In this case, we do not need to freeze the tensor. Instead, we use torch.ops.aten._assert_tensor_metadata.default to ensure that a must not have the same dtype as b. We need to remove ` torch.ops.aten._assert_tensor_metadata.default` for executorch. Update to pin to include pytorch/executorch#7277. Test Plan: ``` buck2 run 'fbcode//mode/dev-nosan' fbcode//caffe2/test:test_export -- -r test_float_conversion buck2 run 'fbcode//mode/dev-nosan' fbcode//caffe2/test:test_export -- -r test_device_to_mutation_float ADD_NEW_STABILITY_CONFIGS=true buck2 run 'fbcode//mode/opt' fbcode//aps_models/ads/modules/module_stability_tests/ctr_cvr_task_arch_stability_tests:inference_stability_tests buck2 run 'fbcode//mode/dev-nosan' fbcode//deeplearning/aot_inductor/cpu/test:cpu_lowering_utils_test buck2 run 'fbcode//mode/dev-nosan' //executorch/exir/tests:test_memory_format_ops_pass buck test aps_models/ads/modules/module_stability_tests/histogram_binning_calibration_tests:calibration_by_feature_stability_test -- --env OVERRIDE_IR_CONFIGS=true buck test aps_models/ads/modules/module_stability_tests/histogram_binning_calibration_tests:calibration_stability_test -- test_stability_config --env OVERRIDE_IR_CONFIGS=true buck2 test 'fbcode//mode/opt' fbcode//aps_models/ads/modules/module_stability_tests/soft_max_loss_with_class_weights:stability_tests -- test_stability_config --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/loss_stability_tests:distillation_stability_tests -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/loss_stability_tests:stability_tests - aps_models.ads.modules.module_stability_tests.loss_stability_tests.ctr_cvr_logistic_regression_loss_stability_test.CtrCvrLogisticRegressionLossStabilityTest: test_cider_loss -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/feature_perturbation_stability_tests:sparse_feature_perturbation_test -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/ctr_cvr_task_arch_stability_tests:train_stability_tests -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/ctr_cvr_task_arch_stability_tests:inference_stability_tests -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/ctr_cvr_task_arch_stability_tests:histogram_binning_stability_test -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/aggregator_stability_tests:replicate_aggregator_stability_test -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/aggregator_stability_tests:label_aggregator_stability_test -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/aggregator_stability_tests:dral_aggregator_stability_test -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/adjust_task_weight_stability_tests:task_weight_stability_test -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/loss_modules/module_stability_tests/cider_stability_tests:stability_tests -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/py2pt/calibration/module_stability_tests/histogram_binning_calibration_stability_tests:stability_tests -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/py2pt/calibration/module_stability_tests/histogram_binning_calibration_by_feature_stability_tests:stability_tests -- --env OVERRIDE_IR_CONFIGS=true ``` Reviewed By: bdhirsh, shruthign Differential Revision: D66988295
pytorch-bot bot
pushed a commit
to pytorch/pytorch
that referenced
this pull request
Dec 20, 2024
Summary: if we have a.to(b), and b has a different dtype with a, then it must be a copy. In this case, we do not need to freeze the tensor. Instead, we use torch.ops.aten._assert_tensor_metadata.default to ensure that a must not have the same dtype as b. We need to remove ` torch.ops.aten._assert_tensor_metadata.default` for executorch. Update to pin to include pytorch/executorch#7277. Test Plan: ``` buck2 run 'fbcode//mode/dev-nosan' fbcode//caffe2/test:test_export -- -r test_float_conversion buck2 run 'fbcode//mode/dev-nosan' fbcode//caffe2/test:test_export -- -r test_device_to_mutation_float ADD_NEW_STABILITY_CONFIGS=true buck2 run 'fbcode//mode/opt' fbcode//aps_models/ads/modules/module_stability_tests/ctr_cvr_task_arch_stability_tests:inference_stability_tests buck2 run 'fbcode//mode/dev-nosan' fbcode//deeplearning/aot_inductor/cpu/test:cpu_lowering_utils_test buck2 run 'fbcode//mode/dev-nosan' //executorch/exir/tests:test_memory_format_ops_pass buck test aps_models/ads/modules/module_stability_tests/histogram_binning_calibration_tests:calibration_by_feature_stability_test -- --env OVERRIDE_IR_CONFIGS=true buck test aps_models/ads/modules/module_stability_tests/histogram_binning_calibration_tests:calibration_stability_test -- test_stability_config --env OVERRIDE_IR_CONFIGS=true buck2 test 'fbcode//mode/opt' fbcode//aps_models/ads/modules/module_stability_tests/soft_max_loss_with_class_weights:stability_tests -- test_stability_config --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/loss_stability_tests:distillation_stability_tests -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/loss_stability_tests:stability_tests - aps_models.ads.modules.module_stability_tests.loss_stability_tests.ctr_cvr_logistic_regression_loss_stability_test.CtrCvrLogisticRegressionLossStabilityTest: test_cider_loss -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/feature_perturbation_stability_tests:sparse_feature_perturbation_test -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/ctr_cvr_task_arch_stability_tests:train_stability_tests -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/ctr_cvr_task_arch_stability_tests:inference_stability_tests -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/ctr_cvr_task_arch_stability_tests:histogram_binning_stability_test -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/aggregator_stability_tests:replicate_aggregator_stability_test -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/aggregator_stability_tests:label_aggregator_stability_test -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/aggregator_stability_tests:dral_aggregator_stability_test -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/adjust_task_weight_stability_tests:task_weight_stability_test -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/loss_modules/module_stability_tests/cider_stability_tests:stability_tests -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/py2pt/calibration/module_stability_tests/histogram_binning_calibration_stability_tests:stability_tests -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/py2pt/calibration/module_stability_tests/histogram_binning_calibration_by_feature_stability_tests:stability_tests -- --env OVERRIDE_IR_CONFIGS=true ``` Reviewed By: bdhirsh, shruthign Differential Revision: D66988295
yushangdi
added a commit
to yushangdi/pytorch
that referenced
this pull request
Jan 3, 2025
Summary: if we have a.to(b), and b has a different dtype with a, then it must be a copy. In this case, we do not need to freeze the tensor. Instead, we use torch.ops.aten._assert_tensor_metadata.default to ensure that a must not have the same dtype as b. We need to remove ` torch.ops.aten._assert_tensor_metadata.default` for executorch. Update to pin to include pytorch/executorch#7277. Test Plan: ``` buck2 run 'fbcode//mode/dev-nosan' fbcode//caffe2/test:test_export -- -r test_float_conversion buck2 run 'fbcode//mode/dev-nosan' fbcode//caffe2/test:test_export -- -r test_device_to_mutation_float ADD_NEW_STABILITY_CONFIGS=true buck2 run 'fbcode//mode/opt' fbcode//aps_models/ads/modules/module_stability_tests/ctr_cvr_task_arch_stability_tests:inference_stability_tests buck2 run 'fbcode//mode/dev-nosan' fbcode//deeplearning/aot_inductor/cpu/test:cpu_lowering_utils_test buck2 run 'fbcode//mode/dev-nosan' //executorch/exir/tests:test_memory_format_ops_pass buck test aps_models/ads/modules/module_stability_tests/histogram_binning_calibration_tests:calibration_by_feature_stability_test -- --env OVERRIDE_IR_CONFIGS=true buck test aps_models/ads/modules/module_stability_tests/histogram_binning_calibration_tests:calibration_stability_test -- test_stability_config --env OVERRIDE_IR_CONFIGS=true buck2 test 'fbcode//mode/opt' fbcode//aps_models/ads/modules/module_stability_tests/soft_max_loss_with_class_weights:stability_tests -- test_stability_config --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/loss_stability_tests:distillation_stability_tests -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/loss_stability_tests:stability_tests - aps_models.ads.modules.module_stability_tests.loss_stability_tests.ctr_cvr_logistic_regression_loss_stability_test.CtrCvrLogisticRegressionLossStabilityTest: test_cider_loss -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/feature_perturbation_stability_tests:sparse_feature_perturbation_test -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/ctr_cvr_task_arch_stability_tests:train_stability_tests -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/ctr_cvr_task_arch_stability_tests:inference_stability_tests -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/ctr_cvr_task_arch_stability_tests:histogram_binning_stability_test -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/aggregator_stability_tests:replicate_aggregator_stability_test -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/aggregator_stability_tests:label_aggregator_stability_test -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/aggregator_stability_tests:dral_aggregator_stability_test -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/module_stability_tests/adjust_task_weight_stability_tests:task_weight_stability_test -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/loss_modules/module_stability_tests/cider_stability_tests:stability_tests -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/py2pt/calibration/module_stability_tests/histogram_binning_calibration_stability_tests:stability_tests -- --env OVERRIDE_IR_CONFIGS=true buck2 test fbcode//aps_models/ads/modules/py2pt/calibration/module_stability_tests/histogram_binning_calibration_by_feature_stability_tests:stability_tests -- --env OVERRIDE_IR_CONFIGS=true ``` Reviewed By: bdhirsh, shruthign Differential Revision: D66988295
yushangdi
added a commit
to yushangdi/pytorch
that referenced
this pull request
Jan 7, 2025
Summary: if we have a.to(b), and b has a different dtype with a, then it must be a copy. In this case, we do not need to freeze the tensor. Instead, we use torch.ops.aten._assert_tensor_metadata.default to ensure that a must not have the same dtype as b. We need to remove ` torch.ops.aten._assert_tensor_metadata.default` for executorch and mtia. Update to executorch pin to include pytorch/executorch#7277. Reviewed By: bdhirsh, shruthign Differential Revision: D66988295
yushangdi
added a commit
to yushangdi/pytorch
that referenced
this pull request
Jan 7, 2025
Summary: if we have a.to(b), and b has a different dtype with a, then it must be a copy. In this case, we do not need to freeze the tensor. Instead, we use torch.ops.aten._assert_tensor_metadata.default to ensure that a must not have the same dtype as b. We need to remove ` torch.ops.aten._assert_tensor_metadata.default` for executorch and mtia. Update to executorch pin to include pytorch/executorch#7277. Reviewed By: bdhirsh, shruthign Differential Revision: D66988295
pytorchmergebot
pushed a commit
to pytorch/pytorch
that referenced
this pull request
Jan 8, 2025
Summary: if we have a.to(b), and b has a different dtype with a, then it must be a copy. In this case, we do not need to freeze the tensor. Instead, we use torch.ops.aten._assert_tensor_metadata.default to ensure that a must not have the same dtype as b. Fixes #139718 Update executorch pin to include pytorch/executorch#7277. Test Plan: ``` buck2 run 'fbcode//mode/dev-nosan' fbcode//caffe2/test:test_export -- -r test_float_conversion buck2 run 'fbcode//mode/dev-nosan' fbcode//caffe2/test:test_export -- -r test_device_to_mutation_float ``` Differential Revision: D66988295 Pull Request resolved: #142420 Approved by: https://github.com/bdhirsh
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.
Summary:
_assert_tensor_metadatanodes is added in D66988295.Differential Revision: D67057219