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
↓ 2 callersFunctioncopy_async
HugeCTR/core23/tensor_operations.cpp:65
↓ 2 callersFunctioncopy_common
HugeCTR/core23/low_level_primitives.cpp:67
↓ 2 callersMethodcopy_non_trainable_params_from_train_layers_to_evaluate_layers
HugeCTR/src/core23_network.cpp:268
↓ 2 callersFunctioncopy_test
test/utest/io/hdfs_backend_test.cpp:105
↓ 2 callersMethodcopy_weights_from_train_layers_to_evaluate_layers
HugeCTR/src/core23_network.cpp:257
↓ 2 callersFunctioncpu_add_bias
test/utest/core23_layer_test/fully_connected_layer_test.cpp:50
↓ 2 callersFunctioncpu_sigmoid
test/utest/core23_layer_test/gru_layer_test.cpp:118
↓ 2 callersFunctioncreate_grouped_embeddings
HugeCTR/embedding/embedding.cpp:24
↓ 2 callersFunctioncross_entropy_loss
test/utest/loss/loss_test.cpp:37
↓ 2 callersFunctiondata_reader_worker_test_impl
test/utest/data_reader/data_reader_parquet_test.cpp:1001
↓ 2 callersMethoddeallocate
* Deallocation function used by std::vector. */
HugeCTR/core/memory.hpp:92
↓ 2 callersMethoddelete_file
HugeCTR/src/io/s3_filesystem.cpp:190
↓ 2 callersFunctiondelete_test
test/utest/io/hdfs_backend_test.cpp:112
↓ 2 callersMethoddense_communicate
HugeCTR/embedding/operators/communication.cpp:75
↓ 2 callersMethoddevice_release_last_batch_here
HugeCTR/src/data_readers/multi_hot/detail/data_reader_impl.cpp:204
↓ 2 callersMethoddo_lookup
(self, embedding_layer, inputs, training)
sparse_operation_kit/documents/tutorials/DenseDemo/models.py:308
↓ 2 callersMethoddownload_params_to_files
HugeCTR/src/pybind/model.cpp:1236
↓ 2 callersMethoddump_by_id
HugeCTR/embedding_storage/dynamic_embedding_cpu.hpp:398
↓ 2 callersMethodembedding_layer
(self)
sparse_operation_kit/documents/tutorials/DLRM/models.py:290
↓ 2 callersFunctionevaluate
(model, dataset, thresholds)
sparse_operation_kit/documents/tutorials/DLRM_Benchmark/trainer.py:9
↓ 2 callersFunctionevaluate
(model, dataset, thresholds)
sparse_operation_kit/SOK_DLRM_Benchmark/trainer.py:10
↓ 2 callersFunctionextract_dir
* Extract directory from path */
HugeCTR/include/data_generator.hpp:68
↓ 2 callersFunctionf2i_input
test/utest/core23_layer_test/masked_softmax_layer_test.cpp:32
↓ 2 callersFunctionf2i_input
test/utest/core23_layer_test/sequence_mask_layer_test.cpp:47
↓ 2 callersFunctionfetch2local_test
test/utest/io/hdfs_backend_test.cpp:119
↓ 2 callersFunctionfile_loader_hdfs_test_impl
test/utest/io/file_loader_test.cpp:24
↓ 2 callersMethodforward
(self, key, bucket_range, embedding_tables, output_buffer)
notebooks/prototype_embedding_collection/main.py:21
↓ 2 callersMethodforward
HugeCTR/embedding/data_parallel_embedding.cpp:204
↓ 2 callersMethodforward_per_gpu
HugeCTR/src/embeddings/embedding_collection.cpp:358
↓ 2 callersMethodfreeze
HugeCTR/include/embeddings/localized_slot_sparse_embedding_hash.hpp:603
↓ 2 callersFunctionfused_reshape_concat_cpu
test/utest/core23_layer_test/fused_reshape_concat_layer_test.cpp:38
↓ 2 callersFunctionfused_reshape_concat_general_cpu
test/utest/core23_layer_test/fused_reshape_concat_general_layer_test.cpp:39
↓ 2 callersMethodgather_nd
(self, *args, **kwargs)
sparse_operation_kit/sparse_operation_kit/dynamic_variable.py:389
↓ 2 callersMethodget_accomplished_workers
HugeCTR/include/data_readers/row_group_reading_thread.hpp:93
↓ 2 callersMethodget_alignment
HugeCTR/src/data_readers/multi_hot/detail/aio_context.cpp:127
↓ 2 callersFunctionget_allreduce_buffer_num_keys
HugeCTR/embedding/common.cpp:635
↓ 2 callersMethodget_batch_size_bytes
HugeCTR/include/data_readers/multi_hot/detail/batch_locations.hpp:83
↓ 2 callersMethodget_buffer
HugeCTR/core23/buffer_channel.hpp:54
↓ 2 callersMethodget_buffer
HugeCTR/include/tensor2.hpp:187
↓ 2 callersFunctionget_byte_per_elem
(args)
benchmarks/embedding_collection/hugectr/sharding/generate_plan.py:23
↓ 2 callersMethodget_cc_major
HugeCTR/include/gpu_resource.hpp:98
↓ 2 callersFunctionget_column_wise_sharding_factor
HugeCTR/include/embeddings/embedding_collection.hpp:83
↓ 2 callersMethodget_column_wise_sharding_nums
(self)
benchmarks/embedding_collection/hugectr/sharding/planner.py:239
↓ 2 callersMethodget_cont_names
HugeCTR/include/data_readers/metadata.hpp:80
↓ 2 callersMethodget_cost
( self, ss: ShardingState, )
samples/dlrm/sharding/planner.py:158
↓ 2 callersMethodget_cost
( self, ss: ShardingState, )
benchmarks/embedding_collection/hugectr/sharding/planner.py:295
↓ 2 callersMethodget_current_num_row_groups
HugeCTR/src/data_readers/row_group_reading_thread.cpp:149
↓ 2 callersMethodget_current_stream_name
HugeCTR/include/gpu_resource.hpp:73
↓ 2 callersMethodget_device_data
HugeCTR/include/data_readers/multi_hot/detail/data_reader_impl.hpp:54
↓ 2 callersMethodget_id
HugeCTR/include/data_readers/multi_hot/detail/data_reader_impl.hpp:79
↓ 2 callersMethodget_label_names
HugeCTR/include/data_readers/metadata.hpp:81
↓ 2 callersMethodget_learning_rate_schedulers
TODO: a workaround to enable GPU LR for HE only; need a better way
HugeCTR/include/embedding.hpp:39
↓ 2 callersMethodget_local_batch_size_bytes
HugeCTR/include/data_readers/multi_hot/detail/data_reader_impl.hpp:81
↓ 2 callersMethodget_local_row_group_id
HugeCTR/src/data_readers/row_group_reading_thread.cpp:158
↓ 2 callersMethodget_loss
HugeCTR/src/core23_network.cpp:92
↓ 2 callersMethodget_max_row_group
HugeCTR/include/data_readers/metadata.hpp:99
↓ 2 callersMethodget_model_comm_buffer_size
(self, batch_size)
notebooks/prototype_embedding_collection/embedding.py:195
↓ 2 callersMethodget_network_comm_buffer_size
(self, batch_size)
notebooks/prototype_embedding_collection/embedding.py:204
↓ 2 callersMethodget_nnz_ptr
HugeCTR/include/sparse_tensor.hpp:123
↓ 2 callersMethodget_non_trainable_params_as_tensors
HugeCTR/include/layer.hpp:67
↓ 2 callersMethodget_opt_state_tensors
HugeCTR/include/optimizer.hpp:246
↓ 2 callersMethodget_parameter_info_from_model
HugeCTR/embedding_storage/weight_io/parameter_IO.cpp:84
↓ 2 callersMethodget_per_class_metric
HugeCTR/include/metrics.hpp:299
↓ 2 callersMethodget_size
test/utest/embedding/cpu_hashtable.hpp:56
↓ 2 callersMethodget_size_in_bytes
HugeCTR/include/general_buffer2.hpp:267
↓ 2 callersMethodget_table_shard_id
HugeCTR/embedding/common.hpp:253
↓ 2 callersFunctionget_tensor_names
(clause)
onnx_converter/hugectr2onnx/hugectr_loader.py:59
↓ 2 callersMethodget_tensorbag
HugeCTR/include/pybind/model.hpp:307
↓ 2 callersMethodget_valid_alignment
HugeCTR/core23/allocator.hpp:35
↓ 2 callersFunctionget_variance
test/utest/core23_layer_test/prelu_dice_layer_test.cpp:48
↓ 2 callersFunctionget_wgrad_max_num_keys
HugeCTR/embedding/common.cpp:608
↓ 2 callersFunctionhas_concat_embedding
(embedding_params: typing.List[EmbeddingParam])
notebooks/prototype_embedding_collection/common.py:26
↓ 2 callersMethodhier_communicate
HugeCTR/embedding/operators/communication.cpp:112
↓ 2 callersFunctionhugectr2onnx_din_test
( batch_size, num_batches, data_source, data_file, graph_config, dense_model, spar
test/onnx_converter_test/hugectr2onnx_din_test.py:24
↓ 2 callersFunctionhybrid_sharding
test/utest/embedding_collection/configuration.hpp:259
↓ 2 callersMethodinit_attr
HugeCTR/embedding/common.cpp:644
↓ 2 callersFunctioninit_optimizer_params
HugeCTR/src/pybind/model.cpp:258
↓ 2 callersMethodinit_sparse_data
test/utest/data_reader/split_batch_test.cpp:44
↓ 2 callersFunctioninitialize_label_and_input
test/utest/network/network_build_test.cpp:89
↓ 2 callersFunctionisResourceVariable
(variable)
sparse_operation_kit/sparse_operation_kit/lookup.py:177
↓ 2 callersFunctionis_mlp_layer
test/utest/core23_layer_test/mlp_test.cpp:254
↓ 2 callersMethodis_started
HugeCTR/src/data_readers/data_reader.cpp:226
↓ 2 callersMethodis_trainable
HugeCTR/include/embeddings/localized_slot_sparse_embedding_hash.hpp:607
↓ 2 callersMethodis_valid_stage
HugeCTR/embedding/data_parallel_embedding.cpp:287
↓ 2 callersFunctionkahanSum
test/utest/embedding_collection/embedding_collection_utils.hpp:93
↓ 2 callersMethodload_dense_optimizer_states
HugeCTR/src/pybind/model.cpp:439
↓ 2 callersMethodload_dense_weights
HugeCTR/src/pybind/model.cpp:479
↓ 2 callersMethodload_metadata
HugeCTR/embedding_storage/weight_io/parameter_IO.cpp:37
↓ 2 callersMethodload_sparse_optimizer_states
HugeCTR/src/pybind/model.cpp:447
↓ 2 callersMethodload_sparse_weights
HugeCTR/src/pybind/model.cpp:488
↓ 2 callersFunctionload_table_to_filesystem_static
(var, optimizer, path)
sparse_operation_kit/sparse_operation_kit/dump_load.py:847
↓ 2 callersMethodmake_shard_matrix
sparse_operation_kit/kit_src/lookup/kernels/embedding_collection.cc:91
↓ 2 callersFunctionmatmul_dgrad_cpu
test/utest/core23_layer_test/matrix_multiply_test.cpp:65
↓ 2 callersFunctionmpi_finalize
test/utest/test_utils.hpp:183
↓ 2 callersFunctionmpi_init
test/utest/test_utils.hpp:172
↓ 2 callersFunctionmulti_head_attention_dgrad_cpu
test/utest/core23_layer_test/multi_head_attention_layer_test.cpp:328
↓ 2 callersFunctionnaiveCoalescedReduction
test/utest/prims/reduce.h:85
↓ 2 callersFunctionnew_delete_resource_manager
test/utest/resource_manager/resource_manager_test.cpp:6
↓ 2 callersFunctionno_print
()
sparse_operation_kit/documents/tutorials/DLRM/utils.py:225
↓ 2 callersFunctionno_print
()
test/sok_perf_test/dlrm/utils.py:225
← previousnext →401–500 of 3,072, ranked by callers