This repository was archived by the owner on Jul 1, 2025. It is now read-only.
Parallelize the EmbeddingBag operator (#27477) #4049
Closed
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
Summary: Pull Request resolved: pytorch/pytorch#27477 We would like to add the intra-op parallelization support for the EmbeddingBag operator. This should bring speedup for the DLRM benchmark: pytorch/pytorch#24385 Benchmark code: ``` from __future__ import absolute_import, division, print_function, unicode_literals import torch import time eb = torch.nn.EmbeddingBag(1000000, 64, mode='sum') input = torch.LongTensor(1500).random_(0, 1000000) offsets = torch.zeros(64, dtype=torch.int64) niter = 10000 s = time.time() for _ in range(niter): out = eb(input, offsets) time_per_iter = (time.time() - s) / niter print('time_per_iter', time_per_iter) print('GB/s', (input.numel() * 64 * 4 + out.numel() * 4) / time_per_iter / 1e9) ``` The following results are single core on Skylake T6: - Before our change (with the original caffe2::EmbeddingLookup) time_per_iter 6.313693523406982e-05 GB/s 6.341517821789133 - After our change using the EmbeddingLookupIdx API which takes the offsets instead of lengths. time_per_iter 5.7627105712890626e-05 GB/s 6.947841559053659 - With Intel's PR: pytorch/pytorch#24385 time_per_iter 7.393271923065185e-05 GB/s 5.415518381664018 For multi-core performance, because Clang doesn't work with OMP, I can only see the single-core performance on SKL T6. ghstack-source-id: 97124557 Differential Revision: D17768404 fbshipit-source-id: 8db0309715ede5ca3084b1f676d763c33e2be6d0
|
This pull request was exported from Phabricator. Differential Revision: D17768404 |
facebook-github-bot
pushed a commit
to pytorch/pytorch
that referenced
this pull request
Jan 24, 2020
Summary: Pull Request resolved: pytorch/glow#4049 Pull Request resolved: #27477 We would like to add the intra-op parallelization support for the EmbeddingBag operator. This should bring speedup for the DLRM benchmark: #24385 Benchmark code: ``` from __future__ import absolute_import, division, print_function, unicode_literals import torch import time eb = torch.nn.EmbeddingBag(1000000, 64, mode='sum') input = torch.LongTensor(1500).random_(0, 1000000) offsets = torch.zeros(64, dtype=torch.int64) niter = 10000 s = time.time() for _ in range(niter): out = eb(input, offsets) time_per_iter = (time.time() - s) / niter print('time_per_iter', time_per_iter) print('GB/s', (input.numel() * 64 * 4 + out.numel() * 4) / time_per_iter / 1e9) ``` The following results are single core on Skylake T6: - Before our change (with the original caffe2::EmbeddingLookup) time_per_iter 6.313693523406982e-05 GB/s 6.341517821789133 - After our change using the EmbeddingLookupIdx API which takes the offsets instead of lengths. time_per_iter 5.7627105712890626e-05 GB/s 6.947841559053659 - With Intel's PR: #24385 time_per_iter 7.393271923065185e-05 GB/s 5.415518381664018 For multi-core performance, because Clang doesn't work with OMP, I can only see the single-core performance on SKL T6. ghstack-source-id: 97124557 Test Plan: With D16990830: ``` buck run mode/dev //caffe2/caffe2/perfkernels:embedding_bench ``` With D17750961: ``` buck run mode/opt //experimental/jianyuhuang/embeddingbag:eb buck run mode/opt-lto //experimental/jianyuhuang/embeddingbag:eb ``` OSS test ``` python run_test.py -i nn -- TestNNDeviceTypeCPU.test_EmbeddingBag_per_sample_weights_and_new_offsets_cpu ``` Buck test ``` buck test mode/dev-nosan //caffe2/test:nn -- "test_EmbeddingBag_per_sample_weights_and_new_offsets_cpu" OMP_NUM_THREADS=3 buck test mode/opt -c pytorch.parallel_backend=tbb //caffe2/test:nn -- "test_EmbeddingBag_per_sample_weights_and_new_offsets" --print-passing-details ``` Generate the AVX2 code for embedding_lookup_idx_avx2.cc: ``` python hp_emblookup_codegen.py --use-offsets ``` Differential Revision: D17768404 fbshipit-source-id: 8dcd15a62d75b737fa97e0eff17f347052675700
|
This pull request has been merged in c9b49cf. |
wuhuikx
pushed a commit
to wuhuikx/pytorch
that referenced
this pull request
Jan 30, 2020
…h#4049) Summary: Pull Request resolved: pytorch/glow#4049 Pull Request resolved: pytorch#27477 We would like to add the intra-op parallelization support for the EmbeddingBag operator. This should bring speedup for the DLRM benchmark: pytorch#24385 Benchmark code: ``` from __future__ import absolute_import, division, print_function, unicode_literals import torch import time eb = torch.nn.EmbeddingBag(1000000, 64, mode='sum') input = torch.LongTensor(1500).random_(0, 1000000) offsets = torch.zeros(64, dtype=torch.int64) niter = 10000 s = time.time() for _ in range(niter): out = eb(input, offsets) time_per_iter = (time.time() - s) / niter print('time_per_iter', time_per_iter) print('GB/s', (input.numel() * 64 * 4 + out.numel() * 4) / time_per_iter / 1e9) ``` The following results are single core on Skylake T6: - Before our change (with the original caffe2::EmbeddingLookup) time_per_iter 6.313693523406982e-05 GB/s 6.341517821789133 - After our change using the EmbeddingLookupIdx API which takes the offsets instead of lengths. time_per_iter 5.7627105712890626e-05 GB/s 6.947841559053659 - With Intel's PR: pytorch#24385 time_per_iter 7.393271923065185e-05 GB/s 5.415518381664018 For multi-core performance, because Clang doesn't work with OMP, I can only see the single-core performance on SKL T6. ghstack-source-id: 97124557 Test Plan: With D16990830: ``` buck run mode/dev //caffe2/caffe2/perfkernels:embedding_bench ``` With D17750961: ``` buck run mode/opt //experimental/jianyuhuang/embeddingbag:eb buck run mode/opt-lto //experimental/jianyuhuang/embeddingbag:eb ``` OSS test ``` python run_test.py -i nn -- TestNNDeviceTypeCPU.test_EmbeddingBag_per_sample_weights_and_new_offsets_cpu ``` Buck test ``` buck test mode/dev-nosan //caffe2/test:nn -- "test_EmbeddingBag_per_sample_weights_and_new_offsets_cpu" OMP_NUM_THREADS=3 buck test mode/opt -c pytorch.parallel_backend=tbb //caffe2/test:nn -- "test_EmbeddingBag_per_sample_weights_and_new_offsets" --print-passing-details ``` Generate the AVX2 code for embedding_lookup_idx_avx2.cc: ``` python hp_emblookup_codegen.py --use-offsets ``` Differential Revision: D17768404 fbshipit-source-id: 8dcd15a62d75b737fa97e0eff17f347052675700
ttumiel
pushed a commit
to ttumiel/pytorch
that referenced
this pull request
Mar 4, 2020
…h#4049) Summary: Pull Request resolved: pytorch/glow#4049 Pull Request resolved: pytorch#27477 We would like to add the intra-op parallelization support for the EmbeddingBag operator. This should bring speedup for the DLRM benchmark: pytorch#24385 Benchmark code: ``` from __future__ import absolute_import, division, print_function, unicode_literals import torch import time eb = torch.nn.EmbeddingBag(1000000, 64, mode='sum') input = torch.LongTensor(1500).random_(0, 1000000) offsets = torch.zeros(64, dtype=torch.int64) niter = 10000 s = time.time() for _ in range(niter): out = eb(input, offsets) time_per_iter = (time.time() - s) / niter print('time_per_iter', time_per_iter) print('GB/s', (input.numel() * 64 * 4 + out.numel() * 4) / time_per_iter / 1e9) ``` The following results are single core on Skylake T6: - Before our change (with the original caffe2::EmbeddingLookup) time_per_iter 6.313693523406982e-05 GB/s 6.341517821789133 - After our change using the EmbeddingLookupIdx API which takes the offsets instead of lengths. time_per_iter 5.7627105712890626e-05 GB/s 6.947841559053659 - With Intel's PR: pytorch#24385 time_per_iter 7.393271923065185e-05 GB/s 5.415518381664018 For multi-core performance, because Clang doesn't work with OMP, I can only see the single-core performance on SKL T6. ghstack-source-id: 97124557 Test Plan: With D16990830: ``` buck run mode/dev //caffe2/caffe2/perfkernels:embedding_bench ``` With D17750961: ``` buck run mode/opt //experimental/jianyuhuang/embeddingbag:eb buck run mode/opt-lto //experimental/jianyuhuang/embeddingbag:eb ``` OSS test ``` python run_test.py -i nn -- TestNNDeviceTypeCPU.test_EmbeddingBag_per_sample_weights_and_new_offsets_cpu ``` Buck test ``` buck test mode/dev-nosan //caffe2/test:nn -- "test_EmbeddingBag_per_sample_weights_and_new_offsets_cpu" OMP_NUM_THREADS=3 buck test mode/opt -c pytorch.parallel_backend=tbb //caffe2/test:nn -- "test_EmbeddingBag_per_sample_weights_and_new_offsets" --print-passing-details ``` Generate the AVX2 code for embedding_lookup_idx_avx2.cc: ``` python hp_emblookup_codegen.py --use-offsets ``` Differential Revision: D17768404 fbshipit-source-id: 8dcd15a62d75b737fa97e0eff17f347052675700
vdantu
pushed a commit
to vdantu/glow
that referenced
this pull request
Jul 12, 2020
Summary: Pull Request resolved: pytorch#4049 Pull Request resolved: pytorch/pytorch#27477 We would like to add the intra-op parallelization support for the EmbeddingBag operator. This should bring speedup for the DLRM benchmark: pytorch/pytorch#24385 Benchmark code: ``` from __future__ import absolute_import, division, print_function, unicode_literals import torch import time eb = torch.nn.EmbeddingBag(1000000, 64, mode='sum') input = torch.LongTensor(1500).random_(0, 1000000) offsets = torch.zeros(64, dtype=torch.int64) niter = 10000 s = time.time() for _ in range(niter): out = eb(input, offsets) time_per_iter = (time.time() - s) / niter print('time_per_iter', time_per_iter) print('GB/s', (input.numel() * 64 * 4 + out.numel() * 4) / time_per_iter / 1e9) ``` The following results are single core on Skylake T6: - Before our change (with the original caffe2::EmbeddingLookup) time_per_iter 6.313693523406982e-05 GB/s 6.341517821789133 - After our change using the EmbeddingLookupIdx API which takes the offsets instead of lengths. time_per_iter 5.7627105712890626e-05 GB/s 6.947841559053659 - With Intel's PR: pytorch/pytorch#24385 time_per_iter 7.393271923065185e-05 GB/s 5.415518381664018 For multi-core performance, because Clang doesn't work with OMP, I can only see the single-core performance on SKL T6. ghstack-source-id: 97124557 Differential Revision: D17768404 fbshipit-source-id: 8dcd15a62d75b737fa97e0eff17f347052675700
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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:
Pull Request resolved: pytorch/pytorch#27477
We would like to add the intra-op parallelization support for the EmbeddingBag operator.
This should bring speedup for the DLRM benchmark:
pytorch/pytorch#24385
Benchmark code:
The following results are single core on Skylake T6:
Before our change (with the original caffe2::EmbeddingLookup)
time_per_iter 6.313693523406982e-05
GB/s 6.341517821789133
After our change using the EmbeddingLookupIdx API which takes the offsets instead of lengths.
time_per_iter 5.7627105712890626e-05
GB/s 6.947841559053659
With Intel's PR: Enhance index_select_add<float> fast path with parallel version pytorch#24385
time_per_iter 7.393271923065185e-05
GB/s 5.415518381664018
For multi-core performance, because Clang doesn't work with OMP, I can only see the single-core performance on SKL T6.
ghstack-source-id: 97124557
Differential Revision: D17768404