pylibwholegraph API

The APIs below are the public WholeGraph Python and PyTorch interfaces in the 26.10 nightly package.

API Documentation

• Initialization
  • pylibwholegraph.torch.init
  • pylibwholegraph.torch.init_torch_env
  • pylibwholegraph.torch.init_torch_env_and_create_wm_comm
  • pylibwholegraph.torch.finalize
• Communicators
  • pylibwholegraph.torch.WholeMemoryCommunicator
  • pylibwholegraph.torch.comm.set_world_info
  • pylibwholegraph.torch.create_group_communicator
  • pylibwholegraph.torch.split_communicator
  • pylibwholegraph.torch.destroy_communicator
  • pylibwholegraph.torch.get_global_communicator
  • pylibwholegraph.torch.get_local_node_communicator
  • pylibwholegraph.torch.get_local_device_communicator
  • pylibwholegraph.torch.get_local_mnnvl_communicator
  • pylibwholegraph.torch.comm.comm_set_distributed_backend
• Tensors
  • pylibwholegraph.torch.WholeMemoryTensor
  • pylibwholegraph.torch.create_wholememory_tensor
  • pylibwholegraph.torch.create_wholememory_tensor_from_filelist
  • pylibwholegraph.torch.destroy_wholememory_tensor
• Embeddings and optimizers
  • pylibwholegraph.torch.WholeMemoryOptimizer
  • pylibwholegraph.torch.create_wholememory_optimizer
  • pylibwholegraph.torch.destroy_wholememory_optimizer
  • pylibwholegraph.torch.WholeMemoryCachePolicy
  • pylibwholegraph.torch.create_wholememory_cache_policy
  • pylibwholegraph.torch.create_builtin_cache_policy
  • pylibwholegraph.torch.destroy_wholememory_cache_policy
  • pylibwholegraph.torch.WholeMemoryEmbedding
  • pylibwholegraph.torch.create_embedding
  • pylibwholegraph.torch.create_embedding_from_filelist
  • pylibwholegraph.torch.destroy_embedding
  • pylibwholegraph.torch.WholeMemoryEmbeddingModule
• Graph storage and operations
  • pylibwholegraph.torch.GraphStructure
  • pylibwholegraph.torch.graph_ops.append_unique
  • pylibwholegraph.torch.graph_ops.add_csr_self_loop
  • pylibwholegraph.torch.wholegraph_ops.unweighted_sample_without_replacement
  • pylibwholegraph.torch.wholegraph_ops.weighted_sample_without_replacement
• Distributed launch
  • pylibwholegraph.torch.add_distributed_launch_options
  • pylibwholegraph.torch.distributed_launch
  • pylibwholegraph.torch.get_rank
  • pylibwholegraph.torch.get_world_size
  • pylibwholegraph.torch.get_local_rank
  • pylibwholegraph.torch.get_local_size
• Training and data-loading helpers
  • pylibwholegraph.torch.add_common_graph_options
  • pylibwholegraph.torch.add_common_model_options
  • pylibwholegraph.torch.add_common_sampler_options
  • pylibwholegraph.torch.add_training_options
  • pylibwholegraph.torch.add_dataloader_options
  • pylibwholegraph.torch.add_node_classfication_options
  • pylibwholegraph.torch.set_framework
  • pylibwholegraph.torch.create_gnn_layers
  • pylibwholegraph.torch.create_sub_graph
  • pylibwholegraph.torch.HomoGNNModel
  • pylibwholegraph.torch.create_node_classification_datasets
  • pylibwholegraph.torch.get_train_dataloader
  • pylibwholegraph.torch.get_valid_test_dataloader
  • pylibwholegraph.torch.compile_cpp_extension
  • pylibwholegraph.torch.get_part_file_name
  • pylibwholegraph.torch.get_part_file_list