Shared numpy ndarray with serialization interface. This class can be used as a drop-in replacement for arguments in multiprocessing. Par

Link prediction network for graphs / knowledge graphs

* @brief General interface of first-order optimizers * * The optimizer class is implemented in a way to trade off compile-time binding and run-time

BlogCatalog social network dataset. Splits: graph, label, train, test Train and test splits are used for link prediction purpos

CIFAR10 dataset for visualization. Splits: train_image_data, train_label_data, test_image_data, test_label_data, image_data, label_d

FB15k knowledge graph dataset. Splits: train, valid, test

FB15k-237 knowledge graph dataset. Splits: train, valid, test

Flickr social network dataset. Splits: graph, label

Freebase knowledge graph dataset. Splits: train

Friendster social network dataset. Splits: graph, small_graph, label

Hyperlink 2012 graph dataset. Splits: pld_train, pld_test

ImageNet dataset for visualization. Splits: train_image, train_feature_data, train_label, train_hierarchical_label, valid_im

MNIST dataset for visualization. Splits: train_image_data, train_label_data, test_image_data, test_label_data, image_data, label_dat

Synthetic math knowledge graph dataset. Splits: train, valid, test

Function call monitor. Parameters: name_style (str): style of displayed function name, can be `full`, `class` or `func`

Node classification network for graphs

WN18 knowledge graph dataset. Splits: train, valid, test

WN18RR knowledge graph dataset. Splits: train, valid, test

Wikidata5m knowledge graph dataset. Splits: train, valid, test

Wikipedia dump for word embedding. Splits: graph

Youtube social network dataset. Splits: graph, label

Helper class for AdaGrad */

Helper class for Adam */

Application(type, *args, **kwargs) Create an application instance of any type. Parameters: type (str): application type,

General interface of graph applications. Parameters: dim (int): dimension of embeddings gpus (list of int, optional): GPU id

Graph dataset. Parameters: name (str): name of dataset urls (dict, optional): url(s) for each split, can be eith

Node embedding application. Given a graph, it embeds each node into a continuous vector representation. The learned embeddings can be us

Knowledge graph embedding application. Given a knowledge graph, it embeds each entity and relation into a continuous vector representation r

* @brief Learning rate schedule * * The learning rate schedule class is used for compatibility of string schedule and custom schedule function. * G

Helper class for Momentum */

Optimizer(type=auto, *args, **kwargs) Create an optimizer instance of any type. Parameters: type (str or auto): optimizer type,

The type of optimizer is used for compile-time binding in GPU code */

Helper class for RMSprop */

Helper class for SGD */

Graph & high-dimensional data visualization. Given a graph or high-dimensional vectors, it maps each node to 2D or 3D coordinates to

Word node embedding application. Given a corpus, it embeds each word into a continuous vector representation. The learned embeddings can

Context manager for working directory. Parameters: dir (str): new working directory

Optimizer interface