libcity.data.dataset.dataset_subclass.gman_dataset

class libcity.data.dataset.dataset_subclass.gman_dataset.GMANDataset(config)

Bases: libcity.data.dataset.traffic_state_point_dataset.TrafficStatePointDataset

get_data_feature()

返回数据集特征，scaler是归一化方法，adj_mx是邻接矩阵，num_nodes是点的个数， feature_dim是输入数据的维度，output_dim是模型输出的维度

Returns

包含数据集的相关特征的字典

Return type

dict

class libcity.data.dataset.dataset_subclass.gman_dataset.Graph(nx_G, is_directed, p, q)

Bases: object

get_alias_edge(src, dst)

Get the alias edge setup lists for a given edge.

node2vec_walk(walk_length, start_node)

Simulate a random walk starting from start node.

preprocess_transition_probs()

Preprocessing of transition probabilities for guiding the random walks.

simulate_walks(num_walks, walk_length)

Repeatedly simulate random walks from each node.

libcity.data.dataset.dataset_subclass.gman_dataset.alias_draw(J, q)

Draw sample from a non-uniform discrete distribution using alias sampling.

libcity.data.dataset.dataset_subclass.gman_dataset.alias_setup(probs)

Compute utility lists for non-uniform sampling from discrete distributions. Refer to https://hips.seas.harvard.edu/blog/2013/03/03/the-alias-method-efficient-sampling-with-many-discrete-outcomes/ for details

libcity.data.dataset.dataset_subclass.gman_dataset.learn_embeddings(walks, dimensions, window_size, iter)