Commit 68fb663
- .circleci/config.yml-53
- .github/scripts/build_wheel.bash+126
- .github/scripts/setup_env.bash+68
- .github/scripts/test_torchrec.bash+102
- .github/scripts/test_wheel.bash+108
- .github/workflows/build_wheel.yml+160
- .github/workflows/fbgemm_nightly_build.yml+23-9
- .github/workflows/fbgemm_nightly_build_cpu.yml-4
- .github/workflows/fbgemm_release_build.yml+3-7
- .github/workflows/fbgemm_release_build_cpu.yml-4
- .github/workflows/fbgemmci.yml+38-14
- .github/workflows/push_wheel_trigger.yml+34
- .github/workflows/test_wheel_trigger.yml+33
- .jenkins/rocm/build_and_test.sh+4-1
- README.md-1
- bench/EmbeddingSpMDMBenchmark.cc+79-5
- fbgemm_gpu/README.md+63-6
- fbgemm_gpu/bench/ssd_table_batched_embeddings_benchmark.py+1-1
- fbgemm_gpu/codegen/embedding_backward_split_template.cu+23-18
- fbgemm_gpu/codegen/embedding_forward_quantized_cpu_template.cpp+40-1
- fbgemm_gpu/codegen/embedding_forward_quantized_host.cpp+10
- fbgemm_gpu/codegen/embedding_forward_quantized_host_cpu.cpp+13
- fbgemm_gpu/codegen/embedding_forward_quantized_split_template.cu+78
- fbgemm_gpu/fbgemm_gpu/_fbgemm_gpu_docs.py+23-7
- fbgemm_gpu/fbgemm_gpu/split_table_batched_embeddings_ops.py+27-7
- fbgemm_gpu/fbgemm_gpu/ssd_split_table_batched_embeddings_ops.py+37-30
- fbgemm_gpu/include/fbgemm_gpu/dispatch_macros.h+28
- fbgemm_gpu/include/fbgemm_gpu/fbgemm_cuda_utils.cuh+4-4
- fbgemm_gpu/include/fbgemm_gpu/sparse_ops.h+17-7
- fbgemm_gpu/include/fbgemm_gpu/split_embeddings_cache_cuda.cuh+3-1
- fbgemm_gpu/setup.py+1-1
- fbgemm_gpu/src/cumem_utils.cu+2
- fbgemm_gpu/src/embedding_inplace_update.cu+6-2
- fbgemm_gpu/src/embedding_inplace_update_cpu.cpp+6-2
- fbgemm_gpu/src/jagged_tensor_ops.cu+148-177
- fbgemm_gpu/src/jagged_tensor_ops_autograd.cpp+116-15
- fbgemm_gpu/src/jagged_tensor_ops_cpu.cpp+28-116
- fbgemm_gpu/src/jagged_tensor_ops_meta.cpp+49
- fbgemm_gpu/src/sparse_ops_gpu.cpp+4-1
- fbgemm_gpu/src/split_embeddings_cache_cuda.cu+42-15
- fbgemm_gpu/src/split_table_batched_embeddings.cpp+1-1
- fbgemm_gpu/test/batched_unary_embeddings_test.py+17
- fbgemm_gpu/test/jagged_tensor_ops_test.py+102-7
- fbgemm_gpu/test/split_table_batched_embeddings_test.py+46-17
- fbgemm_gpu/test/ssd_split_table_batched_embeddings_test.py+7-3
- fbgemm_gpu/test/uvm_test.py+4-1
- include/fbgemm/FbgemmConvert.h-2
- include/fbgemm/FbgemmEmbedding.h+6-3
- include/fbgemm/Types.h+15
- src/EmbeddingSpMDM.cc+126-42
- src/EmbeddingSpMDMAvx2.cc+7-9
- src/RefImplementations.cc+13-15
- src/RefImplementations.h+24-1
- test/EmbeddingSpMDMTest.cc+97-17
- test/EmbeddingSpMDMTestUtils.h+6
0 commit comments