Task¶

`BaseTask`
`NodeClassification`	Node classification tasks.
`LinkPrediction`	Link prediction tasks.
`Recommendation`	Recommendation tasks.
`EdgeClassification`	Edge classification tasks.
`hypergraph`	Hypergraph tasks.

class BaseTask[source]¶

class NodeClassification(args)[source]¶

Node classification tasks.

dataset¶

Task-related dataset

Type: NodeClassificationDataset

evaluator¶

offer evaluation metric

Type: Evaluator

get_graph :: return a graph

get_loss_fn :: return a loss function

class LinkPrediction(args)[source]¶

Link prediction tasks.

dataset¶

Task-related dataset

Type: NodeClassificationDataset

evaluator¶

offer evaluation metric

Type: Evaluator

get_graph :: return a graph

get_loss_fn :: return a loss function

evaluate(n_embedding, r_embedding=None, mode='test')[source]¶

Parameters

n_embedding (th.Tensor) – the embedding of nodes
r_embedding (th.Tensor) – the embedding of relation types
mode (str) – the evaluation mode, train/valid/test

class Recommendation(args)[source]¶: Recommendation tasks.

class EdgeClassification(args)[source]¶

Edge classification tasks.

dataset¶

Task-related dataset

Type: EdgeClassificationDataset

evaluator¶

offer evaluation metric

Type: Evaluator

get_graph :: return a graph

get_loss_fn :: return a loss function

class hypergraph(args)[source]¶

Hypergraph tasks.

dataset¶

Task-related dataset

Type: HypergraphDataset

evaluator¶

offer evaluation metric

Type: Evaluator

get_graph :: return a graph

get_loss_fn :: return a loss function

evaluate(logits, mode='test', info=True)[source]¶

Parameters

logits (th.Tensor) – the logits of batch
mode (str) – the evaluation mode, train/valid/test