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
↓ 1 callersMethod_check_file
(self, label_bin, dense_bin, category_bin)
sparse_operation_kit/documents/tutorials/DLRM_Benchmark/dataset.py:91
↓ 1 callersMethod_check_file
(self, label_bin, dense_bin, category_bin, vocab_sizes)
sparse_operation_kit/documents/tutorials/DLRM_Benchmark/dataset.py:275
↓ 1 callersMethod_check_file
( self, label_bin, dense_bin, category_bin, label_type_byte, sparse_hotness_per_sample )
sparse_operation_kit/SOK_DLRM_Benchmark/dataset.py:102
↓ 1 callersMethod_compute_stats
(self, current_iter: int)
samples/dlrm/mlperf_logger/callbacks.py:44
↓ 1 callersMethod_create_slots_dynamic
(self, var)
sparse_operation_kit/sparse_operation_kit/optimizer.py:126
↓ 1 callersMethod_create_slots_dynamic
(self, var)
sparse_operation_kit/sparse_operation_kit/optimizer.py:263
↓ 1 callersMethod_get
(self, idx)
sparse_operation_kit/documents/tutorials/DLRM_Benchmark/dataset.py:134
↓ 1 callersMethod_get
(self, idx)
sparse_operation_kit/documents/tutorials/DLRM_Benchmark/dataset.py:324
↓ 1 callersMethod_get
(self, idx)
sparse_operation_kit/documents/tutorials/DLRM_Benchmark/dataset.py:396
↓ 1 callersMethod_get
(self, idx)
sparse_operation_kit/SOK_DLRM_Benchmark/dataset.py:154
↓ 1 callersMethod_get_days_for_stage
(self)
samples/dlrm/preprocessing/convert_to_raw.py:72
↓ 1 callersMethod_get_slice_for_stage
(self)
samples/dlrm/preprocessing/convert_to_raw.py:78
↓ 1 callersFunction_hotness_calculate
This function should not be used by user directly.
sparse_operation_kit/sparse_operation_kit/lookup.py:158
↓ 1 callersFunction_infer_step
(tmp_inputs, labels)
sparse_operation_kit/example/sok_train_hps_inference/hps_use_sok_weight_inference.py:182
↓ 1 callersMethod_load_data_for_day
(self, day)
samples/dlrm/preprocessing/convert_to_raw.py:94
↓ 1 callersFunction_lookup_forward
This function should not be used by user directly.
sparse_operation_kit/sparse_operation_kit/lookup.py:204
↓ 1 callersFunction_postprocessing_forward
This function should not be used by user directly.
sparse_operation_kit/sparse_operation_kit/lookup.py:377
↓ 1 callersFunction_preprocessing_forward
This function should not be used by user directly.
sparse_operation_kit/sparse_operation_kit/lookup.py:142
↓ 1 callersFunction_replica_loss
(labels, logits)
sparse_operation_kit/documents/tutorials/DenseDemo/run_sok_horovod.py:69
↓ 1 callersFunction_replica_loss
(labels, logits)
sparse_operation_kit/documents/tutorials/DenseDemo/run_sok_MirroredStrategy.py:59
↓ 1 callersFunction_replica_loss
(labels, logits)
sparse_operation_kit/documents/tutorials/DenseDemo/run_sok_MultiWorker_mpi.py:82
↓ 1 callersFunction_replica_loss
(labels, logits)
sparse_operation_kit/documents/tutorials/DenseDemo/run_tf.py:92
↓ 1 callersFunction_replica_loss
(labels, logits)
sparse_operation_kit/documents/tutorials/MixedPrecision/amp_tf2.py:58
↓ 1 callersFunction_replica_loss
(labels, logits)
sparse_operation_kit/documents/tutorials/MixedPrecision/amp_tf1.py:59
↓ 1 callersFunction_replica_loss
(labels, logits)
test/sok_perf_test/demo/run_tf.py:50
↓ 1 callersMethod_step
(self, samples, labels, first_batch)
sparse_operation_kit/documents/tutorials/DLRM_Benchmark/trainer.py:163
↓ 1 callersMethod_step
(self, samples, labels, first_batch)
sparse_operation_kit/SOK_DLRM_Benchmark/trainer.py:173
↓ 1 callersFunction_train
(model, max_iter, display, max_eval_batches, eval_interval, auc_threshold)
test/pybind_test/model_test.py:1209
↓ 1 callersFunction_train_step
(inputs, labels, first_batch)
sparse_operation_kit/documents/tutorials/DenseDemo/run_sok_horovod.py:74
↓ 1 callersFunction_train_step
(features, labels, first_batch=False)
sparse_operation_kit/documents/tutorials/DLRM/main.py:182
↓ 1 callersFunction_train_step
(inputs, labels)
test/sok_perf_test/demo/run_tf.py:55
↓ 1 callersFunctionada_grad_update_grad
* AdaGrad * ------- * g_i = g_i / s * v_i = v_i + g_i^2 * g_i = -eta * g_i / (sqrt(v_i) + epsilon) */
HugeCTR/embedding_storage/optimizers.hpp:94
↓ 1 callersFunctionadam_update_grad
* Adam * ---- * g_i = g_i / s * m_i = beta_1 * m_i + (1 - beta_1) * g_i * v_i = beta_2 * v_i + (1 - beta_2) * g_i^2 * * m_i_debiased = m_i / (1
HugeCTR/embedding_storage/optimizers.hpp:143
↓ 1 callersFunctionadd_cpu
test/utest/core23_layer_test/add_layer_test.cpp:48
↓ 1 callersFunctionadd_dgrad_cpu
test/utest/core23_layer_test/add_layer_test.cpp:70
↓ 1 callersMethodadd_embedding_collection
HugeCTR/embedding_storage/weight_io/parameter_IO.cpp:33
↓ 1 callersMethodadd_embedding_collection
test/utest/embedding_collection/test_embedding_collection_load_dump.cpp:46
↓ 1 callersMethodadd_layer
Add layer to ONNX graph, one layer may consist of multiple ONNX nodes Args: layer_params: HugeCTR layer parameters wei
onnx_converter/hugectr2onnx/graph_builder.py:36
↓ 1 callersMethodall2all
(self, num_gpus)
notebooks/prototype_embedding_collection/utils.py:11
↓ 1 callersFunctionallSupportedVariableCheckFunc
()
sparse_operation_kit/sparse_operation_kit/lookup.py:195
↓ 1 callersMethodall_p2p_enabled
HugeCTR/src/resource_managers/resource_manager_core.cpp:213
↓ 1 callersMethodallocated
HugeCTR/include/general_buffer2.hpp:263
↓ 1 callersMethodallocated
HugeCTR/include/tensor2.hpp:175
↓ 1 callersMethodallocated
HugeCTR/include/inference/preallocated_buffer2.hpp:49
↓ 1 callersMethodallreduce
(self, num_gpus)
notebooks/prototype_embedding_collection/utils.py:20
↓ 1 callersMethodalltoall
(self, tensor, splits)
sparse_operation_kit/sparse_operation_kit/communication.py:147
↓ 1 callersFunctionanalyze_tensor
HugeCTR/include/parser.hpp:295
↓ 1 callersFunctionappend_test
test/utest/io/local_fs_test.cpp:74
↓ 1 callersFunctionassert_array_eq
test/utest/embedding_collection/embedding_collection_utils.hpp:46
↓ 1 callersMethodbackward_communication_per_gpu
(self)
notebooks/prototype_embedding_collection/embedding.py:314
↓ 1 callersMethodbackward_per_gpu
(self, top_grad, do_allreduce)
notebooks/prototype_embedding_collection/embedding.py:450
↓ 1 callersMethodbackward_per_gpu
HugeCTR/embedding/data_parallel_embedding.cpp:264
↓ 1 callersMethodbackward_per_gpu
HugeCTR/src/embeddings/embedding_collection.cpp:392
↓ 1 callersMethodbatch_fetch
HugeCTR/src/io/s3_filesystem.cpp:267
↓ 1 callersMethodbatch_upload
HugeCTR/src/io/s3_filesystem.cpp:271
↓ 1 callersMethodbias
HugeCTR/include/optimizer.hpp:58
↓ 1 callersMethodbprop
HugeCTR/include/embeddings/all2all_hierarchical.hpp:215
↓ 1 callersFunctionbst
( model: hugectr.Model, optimizer: hugectr.Optimizer_t, )
test/pybind_test/model_test.py:899
↓ 1 callersMethodbuild_s3client
HugeCTR/src/io/s3_filesystem.cpp:108
↓ 1 callersFunctionbytesto
(bytes, to, bsize=1024)
tools/criteo_script/preprocess_nvt.py:70
↓ 1 callersFunctioncal_vocabulary_size_per_gpu_for_distributed_slot
(total_vocabulary_size, num_gpus)
tools/embedding_workspace_calculator/cal_vocabulary_size_per_gpu_and_workspace_size_per_gpu.py:40
↓ 1 callersFunctioncal_vocabulary_size_per_gpu_for_localized_slot
(slot_size_array, num_gpus)
tools/embedding_workspace_calculator/cal_vocabulary_size_per_gpu_and_workspace_size_per_gpu.py:44
↓ 1 callersFunctioncalculate_tensor_dimensions
HugeCTR/src/pybind/add_dense_layer.cpp:937
↓ 1 callersMethodcall
(self, inputs)
sparse_operation_kit/example/sok_train_hps_inference/hps_use_sok_weight_inference.py:127
↓ 1 callersMethodcall
(self, inputs, training=True)
sparse_operation_kit/example/sok_train_hps_inference/sok_train.py:216
↓ 1 callersMethodcapture
HugeCTR/src/graph_wrapper.cpp:25
↓ 1 callersFunctioncast_cpu
test/utest/core23_layer_test/cast_layer_test.cpp:34
↓ 1 callersMethodcheck_destroy
HugeCTR/include/collectives/ib_proxy.hpp:176
↓ 1 callersFunctioncheck_optimizer_is_valid
(optimizer, dump_vars)
sparse_operation_kit/sparse_operation_kit/dump_load.py:363
↓ 1 callersFunctioncheck_overlap
(arr_list, arr2)
sparse_operation_kit/sparse_operation_kit/test/function_test/tf1/lookup/sparse_read_evict.py:25
↓ 1 callersFunctioncheck_overlap
(arr_list, arr2)
sparse_operation_kit/sparse_operation_kit/test/function_test/tf2/lookup/sparse_read_evict.py:25
↓ 1 callersFunctionclear_tensor
(tensor)
notebooks/prototype_embedding_collection/op.py:39
↓ 1 callersFunctioncompare_array
test/utest/core23_layer_test/mlp_test.cpp:129
↓ 1 callersFunctioncompare_array
test/utest/optimizer/optimizer_cpu.hpp:23
↓ 1 callersFunctioncompare_bit_array
test/utest/core23_layer_test/mlp_test.cpp:113
↓ 1 callersFunctioncompare_wgard_dict
test/utest/embedding_collection/reference_embedding.hpp:292
↓ 1 callersMethodcompress_table_ids
HugeCTR/embedding_storage/dynamic_embedding_cpu.hpp:217
↓ 1 callersMethodconnect
HugeCTR/src/io/hadoop_filesystem.cpp:126
↓ 1 callersFunctionconstruct_model
( model_type: ModelType, model: hugectr.Model, optimizer: hugectr.Optimizer_t, single_slot: bool )
test/pybind_test/model_test.py:1189
↓ 1 callersMethodconv_weight_
HugeCTR/src/core23_network.cpp:360
↓ 1 callersFunctionconvert
Convert a HugeCTR model to an ONNX model Args: onnx_model_path: the path to store the ONNX model graph_config: the graph configura
onnx_converter/hugectr2onnx/converter.py:22
↓ 1 callersMethodconvert_func
(self, shard_id)
sparse_operation_kit/documents/tutorials/DLRM/bin2csv.py:98
↓ 1 callersMethodconvert_func
(self, shard_id)
test/sok_perf_test/dlrm/bin2csv.py:98
↓ 1 callersFunctionconvert_pic_to_base64
(picname)
sparse_operation_kit/documents/source/util.py:22
↓ 1 callersFunctionconvert_sparse_tensor_to_core_tensor
HugeCTR/include/core23_helper.hpp:59
↓ 1 callersFunctionconvert_tensor2_to_core23_tensor
HugeCTR/include/core23_helper.hpp:41
↓ 1 callersFunctioncopy_cpu_to_cpu
HugeCTR/core23/low_level_primitives.cpp:30
↓ 1 callersFunctioncopy_cpu_to_gpu
HugeCTR/core23/low_level_primitives.cpp:57
↓ 1 callersFunctioncopy_data_from_cpu
test/utest/core23_layer_test/mlp_test.cpp:260
↓ 1 callersFunctioncopy_gpu_to_cpu
HugeCTR/core23/low_level_primitives.cpp:62
↓ 1 callersFunctioncopy_gpu_to_gpu
HugeCTR/core23/low_level_primitives.cpp:52
↓ 1 callersFunctioncpu_add_bias
test/utest/core23_layer_test/fully_connected_layer_half_test.cpp:43
↓ 1 callersFunctioncpu_add_bias_and_re
test/utest/core23_layer_test/mlp_test.cpp:76
↓ 1 callersFunctioncpu_dotm
test/utest/core23_layer_test/gru_layer_test.cpp:97
↓ 1 callersFunctioncpu_ht
test/utest/core23_layer_test/gru_layer_test.cpp:125
↓ 1 callersFunctioncpu_reverse_add_bias
test/utest/core23_layer_test/fully_connected_layer_half_test.cpp:51
↓ 1 callersFunctioncpu_reverse_add_bias_and_re
test/utest/core23_layer_test/mlp_test.cpp:91
↓ 1 callersFunctioncpu_tanh
test/utest/core23_layer_test/gru_layer_test.cpp:104
↓ 1 callersFunctioncreate_and_save_inference_graph
(args)
sparse_operation_kit/example/sok_train_hps_inference/hps_use_sok_weight_inference.py:160
↓ 1 callersMethodcreate_and_set_optimizer
HugeCTR/src/core23_network.cpp:339
← previousnext →601–700 of 3,072, ranked by callers