1 file changed
+1
-1
lines changed- .github/scripts/fbgemm_gpu_build.bash+13-3
- .github/scripts/fbgemm_gpu_install.bash+8-5
- .github/scripts/fbgemm_gpu_test.bash+8-6
- .github/scripts/nova_postscript.bash+7
- .github/scripts/nova_prescript.bash+3
- .github/scripts/utils_cuda.bash+8
- .github/scripts/utils_pip.bash+62-34
- .github/scripts/utils_pytorch.bash+35
- .github/scripts/utils_rocm.bash+2
- .github/scripts/utils_system.bash+6
- .github/workflows/fbgemm_gpu_ci_cpu.yml+82-6
- .github/workflows/fbgemm_gpu_ci_cuda.yml+17-7
- .github/workflows/fbgemm_gpu_ci_rocm.yml+54-81
- .github/workflows/fbgemm_gpu_docs.yml+4
- .github/workflows/fbgemm_gpu_pip.yml+20-6
- .github/workflows/fbgemm_gpu_release_cpu.yml+14-4
- .github/workflows/fbgemm_gpu_release_cuda.yml+15-5
- fbgemm_gpu/CMakeLists.txt+1
- fbgemm_gpu/bench/merge_embeddings_benchmark.py-1
- fbgemm_gpu/bench/sparse_ops_benchmark.py-1
- fbgemm_gpu/bench/split_table_batched_embeddings_benchmark.py+24-3
- fbgemm_gpu/codegen/embedding_backward_split_cpu_template.cpp+2-2
- fbgemm_gpu/codegen/embedding_bounds_check_host_cpu.cpp+3-1
- fbgemm_gpu/codegen/embedding_forward_quantized_cpu_template.cpp+90-170
- fbgemm_gpu/codegen/embedding_forward_split_cpu.cpp+42
- fbgemm_gpu/codegen/embedding_forward_split_meta_template.cpp+4
- fbgemm_gpu/fbgemm_gpu/sparse_ops.py+28-14
- fbgemm_gpu/include/fbgemm_gpu/dispatch_macros.h+2
- fbgemm_gpu/setup.py+5-6
- fbgemm_gpu/src/jagged_tensor_ops/jagged_tensor_ops_cpu.cpp+3
- fbgemm_gpu/src/jagged_tensor_ops/jagged_tensor_ops_meta.cpp-26
- fbgemm_gpu/src/memory_utils/memory_utils_ops.cpp-3
- fbgemm_gpu/src/memory_utils/memory_utils_ops_cpu.cpp+26
fbgemm_gpu/src/permute_pooled_embedding_ops/permute_pooled_embedding_ops_cpu.cpp
+2
-1
fbgemm_gpu/src/sparse_ops/common.cuh
+1
-1
fbgemm_gpu/src/sparse_ops/sparse_ops_cpu.cpp
+2
-1
fbgemm_gpu/src/split_embeddings_utils/split_embeddings_utils.cpp
+29
-1
fbgemm_gpu/test/batched_unary_embeddings_test.py
+13
-3
fbgemm_gpu/test/failures_dict.json
+13
-34
fbgemm_gpu/test/failures_dict_fast.json
+13
-330
fbgemm_gpu/test/input_combine_test.py
-4
fbgemm_gpu/test/jagged_tensor_ops_test.py
+25
-28
fbgemm_gpu/test/permute_pooled_embedding_test.py
+46
-7
fbgemm_gpu/test/quantize_ops_test.py
+2
-5
fbgemm_gpu/test/sparse_ops_test.py
+12
-6
fbgemm_gpu/test/split_table_batched_embeddings_test.py
+8
-94
src/DirectConv.h
+1
-1
0 commit comments