path	# lines of code
bench/EmbeddingQuantizeBenchmark.cc	89
bench/TransposeBenchmark.cc	49
bench/GEMMsTunableBenchmark.cc	286
bench/Depthwise3DBenchmark.cc	188
bench/BenchUtils.cc	182
bench/RequantizeBenchmark.cc	123
bench/BenchUtils.h	415
bench/GEMMsBenchmark.cc	260
bench/PackedRequantizeAcc16Benchmark.cc	359
bench/SparseDenseMMFP32Benchmark.cc	93
bench/SparseDenseMMInt8Benchmark.cc	112
bench/I8SpmdmBenchmark.cc	172
bench/PackedRequantizeAcc32Benchmark.cc	262
bench/DepthwiseBenchmark.cc	305
bench/EmbeddingIndexRemappingBenchmark.cc	134
bench/AlignedVec.h	86
bench/FP16Benchmark.cc	40
bench/I64Benchmark.cc	115
bench/EmbeddingSpMDMNBitBenchmark.cc	278
bench/ConvertBenchmark.cc	51
bench/RowOffsetBenchmark.cc	37
bench/ConvUnifiedBenchmark.cc	449
bench/GroupwiseConvRequantizeBenchmark.cc	413
bench/Im2ColFusedRequantizeBenchmark.cc	296
bench/RowwiseAdagradFusedBenchmark.cc	162
bench/RowwiseAdagradBenchmark.cc	189
bench/EmbeddingSpMDMBenchmark.cc	293
bench/PackedFloatInOutBenchmark.cc	242
bench/EmbeddingSpMDMNBitRowWiseSparseBenchmark.cc	309
bench/EmbeddingSpMDM8BitBenchmark.cc	282
bench/SparseAdagradBenchmark.cc	190
fbgemm_gpu/bench/split_embeddings_cache_benchmark.py	448
fbgemm_gpu/bench/jagged_tensor_benchmark.py	58
fbgemm_gpu/bench/merge_embeddings_benchmark.py	440
fbgemm_gpu/bench/batched_unary_embeddings_benchmark.py	137
fbgemm_gpu/bench/bench_utils.py	34
fbgemm_gpu/bench/split_table_batched_embeddings_benchmark.py	1916
fbgemm_gpu/bench/scripts/batch_benchmark_run.py	73
fbgemm_gpu/bench/verify_fp16_stochastic_benchmark.cu	287
fbgemm_gpu/bench/sparse_ops_benchmark.py	52
fbgemm_gpu/bench/quantize_ops_benchmark.py	186
fbgemm_gpu/bench/histogram_binning_calibration_benchmark.py	242
fbgemm_gpu/fbgemm_gpu/split_embedding_configs.py	88
fbgemm_gpu/fbgemm_gpu/permute_pooled_embedding_modules.py	51
fbgemm_gpu/fbgemm_gpu/enums.py	10
fbgemm_gpu/fbgemm_gpu/split_embedding_inference_converter.py	145
fbgemm_gpu/fbgemm_gpu/__init__.py	4
fbgemm_gpu/fbgemm_gpu/uvm.py	19
fbgemm_gpu/fbgemm_gpu/batched_unary_embeddings_ops.py	61
fbgemm_gpu/fbgemm_gpu/split_table_batched_embeddings_ops.py	1963
fbgemm_gpu/version.py	1
fbgemm_gpu/include/fbgemm_gpu/input_combine.h	16
fbgemm_gpu/include/fbgemm_gpu/cub_namespace_postfix.cuh	14
fbgemm_gpu/include/fbgemm_gpu/layout_transform_ops.cuh	105
fbgemm_gpu/include/fbgemm_gpu/cpu_utils.h	42
fbgemm_gpu/include/fbgemm_gpu/embedding_common.h	54
fbgemm_gpu/include/fbgemm_gpu/quantize_ops_utils.h	55
fbgemm_gpu/include/fbgemm_gpu/split_embeddings_utils.cuh	139
fbgemm_gpu/include/fbgemm_gpu/cub_namespace_prefix.cuh	12
fbgemm_gpu/include/fbgemm_gpu/quantize_ops_gpu.h	14
fbgemm_gpu/include/fbgemm_gpu/embedding_backward_template_helpers.cuh	98
fbgemm_gpu/include/fbgemm_gpu/sparse_ops.cuh	35
fbgemm_gpu/include/fbgemm_gpu/quantize_ops.cuh	27
fbgemm_gpu/include/fbgemm_gpu/cuda_utils.cuh	14
fbgemm_gpu/include/fbgemm_gpu/fbgemm_cuda_utils.cuh	1948
fbgemm_gpu/include/fbgemm_gpu/permute_pooled_embs_function.h	1
fbgemm_gpu/include/fbgemm_gpu/permute_pooled_embedding_ops.h	16
fbgemm_gpu/include/fbgemm_gpu/enum_utils.h	57
fbgemm_gpu/include/fbgemm_gpu/merge_pooled_embeddings.h	7
fbgemm_gpu/include/fbgemm_gpu/sparse_ops.h	317
fbgemm_gpu/include/fbgemm_gpu/bench_utils.cuh	65
fbgemm_gpu/include/fbgemm_gpu/sparse_ops_utils.h	214
fbgemm_gpu/include/fbgemm_gpu/dispatch_macros.h	141
fbgemm_gpu/pyproject.toml	18
fbgemm_gpu/codegen/embedding_bounds_check.cu	138
fbgemm_gpu/codegen/embedding_backward_split_host_cpu_template.cpp	200
fbgemm_gpu/codegen/embedding_backward_dense_host.cpp	363
fbgemm_gpu/codegen/embedding_backward_split_template.cu	1085
fbgemm_gpu/codegen/embedding_forward_split_cpu.cpp	526
fbgemm_gpu/codegen/embedding_backward_split_cpu_approx_template.cpp	191
fbgemm_gpu/codegen/embedding_backward_split_host_template.cpp	467
fbgemm_gpu/codegen/embedding_forward_quantized_host_cpu.cpp	380
fbgemm_gpu/codegen/embedding_backward_split_indice_weights_template.cu	274
fbgemm_gpu/codegen/embedding_bounds_check_host_cpu.cpp	98
fbgemm_gpu/codegen/embedding_bounds_check_host.cpp	16
fbgemm_gpu/codegen/embedding_forward_quantized_split_template.cu	917
fbgemm_gpu/codegen/embedding_forward_quantized_cpu_template.cpp	359
fbgemm_gpu/codegen/embedding_forward_split_template.cu	450
fbgemm_gpu/codegen/embedding_forward_split_cpu.h	58
fbgemm_gpu/codegen/embedding_backward_dense_host_cpu.cpp	164
fbgemm_gpu/codegen/embedding_forward_quantized_host.cpp	180
fbgemm_gpu/codegen/embedding_backward_code_generator.py	531
fbgemm_gpu/codegen/embedding_forward_template_helpers.cuh	29
fbgemm_gpu/codegen/embedding_backward_split_cpu_template.cpp	340
fbgemm_gpu/codegen/lookup_args.py	39
fbgemm_gpu/setup.py	144
fbgemm_gpu/src/histogram_binning_calibration_ops.cu	388
fbgemm_gpu/src/split_table_batched_embeddings.cpp	101
fbgemm_gpu/src/sparse_ops_gpu.cpp	157
fbgemm_gpu/src/sparse_ops.cu	1953
fbgemm_gpu/src/input_combine_cpu.cpp	278
fbgemm_gpu/src/sparse_ops_cpu.cpp	1683
fbgemm_gpu/src/layout_transform_ops_cpu.cpp	63
fbgemm_gpu/src/quantize_ops.cu	666
fbgemm_gpu/src/cumem_utils.cu	305
fbgemm_gpu/src/merge_pooled_embeddings_gpu.cpp	341
fbgemm_gpu/src/cumem_utils_host.cpp	34
fbgemm_gpu/src/merge_pooled_embeddings_cpu.cpp	33
fbgemm_gpu/src/jagged_tensor_ops_cpu.cpp	967
fbgemm_gpu/src/layout_transform_ops_gpu.cpp	16
fbgemm_gpu/src/jagged_tensor_ops.cu	1150
fbgemm_gpu/src/split_embeddings_cache_cuda.cu	1803
fbgemm_gpu/src/permute_pooled_embedding_ops_cpu.cpp	1
fbgemm_gpu/src/cumem_utils.h	18
fbgemm_gpu/src/quantize_ops_cpu.cpp	303
fbgemm_gpu/src/quantize_ops_gpu.cpp	35
fbgemm_gpu/src/permute_pooled_embedding_ops_gpu.cpp	132
fbgemm_gpu/src/split_embeddings_utils.cu	284
fbgemm_gpu/src/cpu_utils.cpp	145
fbgemm_gpu/src/layout_transform_ops.cu	129
fbgemm_gpu/src/permute_pooled_embedding_ops.cu	74
cmake/modules/FindMKL.cmake	252
cmake/modules/FindSphinx.cmake	8
include/fbgemm/ConvUtils.h	161
include/fbgemm/Utils.h	214
include/fbgemm/FbgemmSparse.h	135
include/fbgemm/QuantUtilsAvx512.h	21
include/fbgemm/FbgemmFPCommon.h	190
include/fbgemm/spmmUtilsAvx2.h	28
include/fbgemm/FbgemmConvert.h	56
include/fbgemm/spmmUtils.h	44
include/fbgemm/Types.h	116
include/fbgemm/FbgemmBuild.h	60
include/fbgemm/FbgemmI8DepthwiseAvx2.h	63
include/fbgemm/FbgemmI64.h	18
include/fbgemm/FbgemmEmbedding.h	234
include/fbgemm/PackingTraits-inl.h	298
include/fbgemm/QuantUtils.h	224
include/fbgemm/QuantUtilsAvx2.h	111
include/fbgemm/FbgemmI8Spmdm.h	77
include/fbgemm/OutputProcessing-inl.h	282
include/fbgemm/UtilsAvx2.h	51
include/fbgemm/FbgemmPackMatrixB.h	200
include/fbgemm/Fbgemm.h	804
include/fbgemm/FbgemmFP16.h	40
include/fbgemm/FbgemmI8DirectconvAvx2.h	33
defs.bzl	118
src/QuantUtilsAvx512.cc	356
src/FbgemmI8DepthwisePerChannelQuantAvx2.cc	91
src/ExecuteKernel.cc	1
src/QuantUtils.cc	706
src/GenerateKernelDirectConvU8S8S32ACC32.cc	491
src/UtilsAvx2.cc	269
src/spmmUtilsAvx2.cc	271
src/FbgemmSparseDenseInt8Avx512.cc	506
src/RowWiseSparseAdagradFused.cc	823
src/FbgemmSparseDenseInt8Avx2.cc	221
src/GenerateKernelU8S8S32ACC16Avx512.cc	256
src/FbgemmI8Depthwise3DAvx2.cc	989
src/FbgemmFP16UKernelsAvx2.cc	886
src/OptimizedKernelsAvx2.cc	299
src/FbgemmFP16UKernelsAvx512_256.cc	2247
src/PackWeightsForDirectConv.cc	412
src/FbgemmI8Spmdm.cc	296
src/FbgemmSparseDenseAvx512.cc	130
src/TransposeUtilsAvx2.h	343
src/FbgemmI8Depthwise2DAvx2-inl.h	656
src/GenerateKernel.h	99
src/spmmUtils.cc	292
src/GroupwiseConvAcc32Avx512.cc	206
src/FbgemmBfloat16Convert.cc	56
src/FbgemmFP16UKernelsAvx512.cc	3130
src/ExecuteKernelU8S8.h	50
src/PackWeightMatrixForGConv.cc	189
src/Utils.cc	397
src/GenerateI8Depthwise.h	30
src/EmbeddingSpMDM.cc	1251
src/PackBMatrix.cc	264
src/FbgemmFP16UKernelsAvx2.h	14
src/QuantUtilsAvx2.cc	1761
src/FbgemmBfloat16ConvertAvx2.cc	42
src/FbgemmFloat16ConvertAvx512.cc	54
src/FbgemmConv.cc	434
src/FbgemmFP16UKernelsIntrinsicAvx2.cc	86
src/FbgemmFP16UKernelsAvx512.h	22
src/FbgemmBfloat16ConvertAvx512.cc	38
src/SparseAdagrad.cc	842
src/OptimizedKernelsAvx2.h	19
src/Fbgemm.cc	510
src/FbgemmFPCommon.cc	260
src/PackDepthwiseConvMatrixAvx2.cc	113
src/FbgemmFP16UKernelsIntrinsicAvx512_256.cc	92
src/GenerateKernelU8S8S32ACC16Avx512VNNI.cc	27
src/FbgemmFloat16ConvertAvx2.cc	54
src/ExecuteKernelU8S8.cc	486
src/GenerateKernelU8S8S32ACC32Avx512VNNI.cc	271
src/CodeGenHelpers.h	167
src/GenerateI8Depthwise.cc	483
src/FbgemmFloat16Convert.cc	74
src/GenerateKernelU8S8S32ACC32.cc	308
src/GenerateKernelU8S8S32ACC16.cc	262
src/FbgemmI8DepthwiseAvx2.cc	158
src/RefImplementations.cc	1755
src/GroupwiseConvAcc32Avx2.cc	204
src/PackWeightsForConv.cc	172
src/EmbeddingSpMDMAvx512.cc	521
src/FbgemmFP16UKernelsIntrinsicAvx512.cc	111
src/codegen_fp16fp32.cc	461
src/RefImplementations.h	259
src/PackAWithRowOffset.cc	201
src/ExecuteKernel.h	5
src/FbgemmSparseDenseVectorInt8Avx512.cc	217
src/GroupwiseConv.h	256
src/GroupwiseConv.cc	938
src/EmbeddingSpMDMNBit.cc	1170
src/FbgemmI64.cc	409
src/PackAWithIm2Col.cc	645
src/EmbeddingSpMDMAvx2.cc	132
src/CodeCache.h	96
src/FbgemmFP16UKernelsAvx512_256.h	16
src/UtilsAvx512.cc	845
src/FbgemmSparseDense.cc	253
src/TransposeUtils.cc	74
src/FbgemmFP16.cc	119
src/PackAWithQuantRowOffset.cc	217
src/PackAMatrix.cc	153
src/ExecuteKernelGeneric.h	49
src/FbgemmSparseDenseAvx2.cc	117
src/DirectConv.h	148
src/PackMatrix.cc	116
src/MaskAvx2.h	39
src/TransposeUtils.h	31
src/GenerateKernel.cc	15
src/FbgemmI8DepthwiseAvx2-inl.h	487