MCPcopy Create free account

hub / github.com/NVIDIA-Merlin/HugeCTR / functions

Functions3,072 in github.com/NVIDIA-Merlin/HugeCTR

⨍ Functions 3,072◇ Types & classes 981↳ Endpoints 3
↓ 19 callersMethodtest
test/utest/core23_layer_test/multi_cross_layer_test.cpp:631
↓ 18 callersMethodget_ptr
HugeCTR/include/data_readers/source.hpp:37
↓ 18 callersMethodstream
sparse_operation_kit/kit_src/lookup/impl/core_impl/compat/gpu_device_context.h:50
↓ 17 callersMethodload
HugeCTR/include/pybind/training_callback.hpp:42
↓ 17 callersFunctionnum_gpus
()
sparse_operation_kit/sparse_operation_kit/communication.py:225
↓ 17 callersMethodprint
HugeCTR/core23/logger.cpp:51
↓ 17 callersMethodwrite
HugeCTR/include/data_generator.hpp:183
↓ 16 callersFunctionHCTR_LOG_ARGS
HugeCTR/include/common.hpp:227
↓ 16 callersFunctionallgather
(*args, **kwargs)
sparse_operation_kit/sparse_operation_kit/communication.py:240
↓ 16 callersMethodelapsedMilliseconds
HugeCTR/include/utils.hpp:159
↓ 16 callersMethodget_wgrads
HugeCTR/include/trainable_layer.hpp:158
↓ 16 callersMethodsave
(self)
samples/dlrm/preprocessing/convert_to_raw.py:125
↓ 16 callersFunctiontranspose
test/utest/core23_layer_test/multi_head_attention_layer_test.cpp:135
↓ 15 callersMethodalignment
HugeCTR/core23/tensor_params.hpp:60
↓ 15 callersMethodget_global_gpu_count
HugeCTR/include/resource_managers/resource_manager_core.hpp:83
↓ 15 callersMethodreset
* To reset the CSR23 buffer. * You need to call reset when you want to overwrite the original data in this CSR23. */
HugeCTR/include/data_readers/csr23.hpp:144
↓ 15 callersMethodwait_event
HugeCTR/src/pipeline.cpp:58
↓ 14 callersFunctionGetBlobsBufferChannel
HugeCTR/src/network_buffer_channels.cpp:49
↓ 14 callersFunctioncheck_make_dir
* Check if file path exist if not create it. */
HugeCTR/include/data_generator.hpp:54
↓ 14 callersMethodcheck_out_tensor
HugeCTR/src/pybind/model.cpp:1461
↓ 14 callersMethodclear
(self)
notebooks/prototype_embedding_collection/utils.py:33
↓ 14 callersFunctionfill_data
test/utest/core23_layer_test/mlp_test.cpp:48
↓ 14 callersMethodget_local_gpu_count
HugeCTR/include/resource_managers/resource_manager_core.hpp:82
↓ 14 callersMethodrecord_done
HugeCTR/src/pipeline.cpp:65
↓ 14 callersFunctionwrite_file_head
(path, sok_var_info, var_type, var_name, data_index)
sparse_operation_kit/sparse_operation_kit/dump_load.py:337
↓ 13 callersFunctiondata_reader_group_epoch_strided_batch_test_impl
test/utest/data_reader/data_reader_parquet_test.cpp:628
↓ 13 callersMethoddistribute
HugeCTR/include/data_readers/multi_hot/detail/batch_locations.hpp:85
↓ 13 callersMethodget_slot_names
(self)
sparse_operation_kit/sparse_operation_kit/optimizer.py:152
↓ 13 callersMethodhas_table_shard
HugeCTR/embedding/common.hpp:247
↓ 13 callersMethodlookup
( self, keys: typing.List[int], num_keys: int, id_space_offset: typing.List[in
notebooks/prototype_embedding_collection/embedding_table.py:19
↓ 13 callersMethodmake_core_resource
sparse_operation_kit/kit_src/lookup/kernels/embedding_collection.cc:81
↓ 13 callersMethodrun
HugeCTR/src/pipeline.cpp:83
↓ 12 callersFunctionactivate_tensor
HugeCTR/include/parser.hpp:303
↓ 12 callersMethodelapsedSeconds
HugeCTR/include/utils.hpp:161
↓ 12 callersMethodget_full_batchsize
HugeCTR/src/data_readers/data_reader.cpp:218
↓ 12 callersMethodget_nnz_ptr
HugeCTR/include/tensor2.hpp:320
↓ 12 callersMethodget_num
HugeCTR/include/inference_key_generator.hpp:50
↓ 12 callersMethodgraph_to_json
HugeCTR/src/pybind/model.cpp:382
↓ 12 callersFunctionmatmul_cpu
test/utest/core23_layer_test/multi_head_attention_layer_test.cpp:113
↓ 12 callersMethodplan
(self)
benchmarks/embedding_collection/hugectr/sharding/planner.py:484
↓ 12 callersMethodshard
HugeCTR/include/embeddings/embedding_collection.hpp:186
↓ 12 callersMethodstop
HugeCTR/src/data_readers/data_collector.cpp:87
↓ 12 callersMethodwrite_to
HugeCTR/embedding_storage/weight_io/fs_interface.cpp:25
↓ 11 callersFunctioncheck_comm_tool
()
sparse_operation_kit/sparse_operation_kit/communication.py:188
↓ 11 callersMethoddata
HugeCTR/core23/tensor_container.hpp:139
↓ 11 callersMethodgather
(self, tensor, axis)
sparse_operation_kit/documents/tutorials/DLRM/utils.py:144
↓ 11 callersMethodget_batch_size
HugeCTR/include/regularizer.hpp:72
↓ 11 callersMethodget_tf_dtype_by_index
(cls, index)
sparse_operation_kit/sparse_operation_kit/dump_load.py:108
↓ 11 callersMethodmy_params
HugeCTR/core23/tensor.cpp:59
↓ 11 callersMethodname
HugeCTR/core23/device.cpp:47
↓ 11 callersFunctionread_file_head
(path)
sparse_operation_kit/sparse_operation_kit/dump_load.py:349
↓ 11 callersMethodset_stream
HugeCTR/src/pipeline.cpp:45
↓ 11 callersMethodstream
HugeCTR/core23/cuda_stream.hpp:52
↓ 10 callersFunctiondata_reader_group_epoch_squential_batch_test_impl
test/utest/data_reader/data_reader_parquet_test.cpp:842
↓ 10 callersMethodget_value_tensor
HugeCTR/include/tensor2.hpp:302
↓ 10 callersMethodmu
sparse_operation_kit/kit_src/variable/kernels/dummy_var.cc:37
↓ 10 callersMethodrun
(self, func, *args, **kwargs)
test/sok_perf_test/dlrm/utils.py:141
↓ 10 callersFunctionsanity_check
(shard_matrix, shard_strategy)
benchmarks/embedding_collection/hugectr/sharding/test_sharding_planner.py:75
↓ 10 callersMethodsize
HugeCTR/core23/details/tensor_impl.hpp:49
↓ 9 callersMethodallocate
HugeCTR/core23/buffer.cpp:78
↓ 9 callersMethodconvert
test/utest/embedding/sparse_embedding_hash_cpu.hpp:45
↓ 9 callersMethodcount
HugeCTR/core23/device.cpp:34
↓ 9 callersFunctioncpu_mm
test/utest/core23_layer_test/gru_layer_test.cpp:80
↓ 9 callersMethodget_dense_tensor23s
HugeCTR/src/data_readers/multi_hot/async_data_reader.cpp:494
↓ 9 callersMethodget_global_id
* Get global id according to local id. * @param local_id local id. * @return global id, if cannot find this local device in current node return
HugeCTR/include/device_map.hpp:134
↓ 9 callersMethodget_label_tensor23s
HugeCTR/src/data_readers/multi_hot/async_data_reader.cpp:498
↓ 9 callersFunctionglobal_gpu_id
()
sparse_operation_kit/sparse_operation_kit/communication.py:220
↓ 9 callersMethodput
HugeCTR/embedding_storage/dynamic_embedding_cpu.hpp:44
↓ 9 callersFunctionreader_test
test/utest/data_reader/data_reader_v2_async_test.cpp:34
↓ 9 callersFunctionsingle_buffer_test_impl
test/utest/core23/basic_buffer_test.cpp:48
↓ 9 callersMethodstart
HugeCTR/include/utils.hpp:151
↓ 8 callersFunctionSOK_IndexedSlices
()
sparse_operation_kit/sparse_operation_kit/utils.py:7
↓ 8 callersFunctioncheck_cpu_gpu
test/utest/core23_layer_test/fully_connected_layer_test.cpp:26
↓ 8 callersFunctioncompare_array
test/utest/embedding/embedding_test_utils.hpp:46
↓ 8 callersFunctioncopy_test_impl
test/utest/core23/low_level_primitive_test.cpp:61
↓ 8 callersMethodexec_proxy_cmd
HugeCTR/src/collectives/ib_proxy.cpp:890
↓ 8 callersMethodget_device_id
sparse_operation_kit/kit_src/lookup/impl/core_impl/tf_backend.hpp:53
↓ 8 callersMethodget_np_dtype_by_index
(cls, index)
sparse_operation_kit/sparse_operation_kit/dump_load.py:117
↓ 8 callersFunctionget_num_elements_from_dimensions
HugeCTR/include/tensor2.hpp:26
↓ 8 callersMethodget_process_id
HugeCTR/include/resource_managers/resource_manager_core.hpp:87
↓ 8 callersMethodhas_bucket_and_object
HugeCTR/include/io/gcs_filesystem.hpp:49
↓ 8 callersMethodrowoffset_count
HugeCTR/include/tensor2.hpp:322
↓ 8 callersMethodset_device
HugeCTR/include/utils.hpp:228
↓ 8 callersMethodsplit_hot_shard
split the shard with the largest hotness
benchmarks/embedding_collection/hugectr/sharding/planner.py:95
↓ 8 callersMethodsubmit
HugeCTR/src/thread_pool.cpp:79
↓ 8 callersMethodview
HugeCTR/core23/tensor.hpp:79
↓ 7 callersMethodallgather
(self, tensor)
sparse_operation_kit/sparse_operation_kit/communication.py:162
↓ 7 callersMethodallocator_params
HugeCTR/core23/tensor_params.hpp:72
↓ 7 callersFunctioncheck_data_cpu_and_gpu
test/utest/core23_layer_test/mlp_test.cpp:286
↓ 7 callersFunctiondata_reader_group_iter_strided_batch_test_impl
test/utest/data_reader/data_reader_parquet_test.cpp:310
↓ 7 callersMethodget_current_batchsize
HugeCTR/include/data_readers/data_reader.hpp:94
↓ 7 callersMethodget_num_process
from ResourceManager
HugeCTR/include/resource_managers/resource_manager_core.hpp:86
↓ 7 callersMethodget_rowoffset_tensor
HugeCTR/include/tensor2.hpp:310
↓ 7 callersMethodis_unique
HugeCTR/core23/tensor.cpp:63
↓ 7 callersFunctionjoin
HugeCTR/src/pybind/model_compile.cpp:38
↓ 7 callersMethodown_data
HugeCTR/core23/tensor.cpp:65
↓ 7 callersMethodread_a_batch_to_device
HugeCTR/src/data_readers/data_reader.cpp:185
↓ 7 callersFunctionround_up
HugeCTR/include/data_readers/multi_hot/detail/batch_locations.hpp:27
↓ 7 callersMethodsplit_oom_shard
(self, table_id, cost, is_column_wise=False)
benchmarks/embedding_collection/hugectr/sharding/planner.py:174
↓ 7 callersMethodstart
HugeCTR/src/data_readers/data_reader.cpp:230
← previousnext →101–200 of 3,072, ranked by callers