Sampling

class netin.sampling.Sampling(g: Graph, pseeds: float, max_tries: int = 20, random_seed: object | None = None, **kwargs)

Base class for sampling methods.

Parameters

g: netin.Graph | netin.DiGraph

global network

pseeds: float

fraction of seeds to sample

max_tries: int

maximum number of tries to sample a subgraph with enough classes and edges

random_seed: object

seed for random number generator

kwargs: dict

additional parameters for the sampling method

Notes

• The original graph g (passed as parameter) is not modified.

  The sampling method creates a copy of it, and stores it in self.g.

• This class does not create a subgraph.

info()

Prints a summary of the training sample subgraph, including its attributes.

property method_name: str

Name of sampling method.

sampling()

Creates a new instance of the respective sampling method, and calls its respective extract_subgraph method.

class netin.sampling.RandomNodes(g: Graph, pseeds: float, max_tries: int = 20, random_seed: object | None = None, **kwargs)

Random node sampling.

Parameters

g: netin.Graph | netin.DiGraph

global network

pseeds: float

fraction of seeds to sample

max_tries: int

maximum number of tries to sample a subgraph with enough classes and edges

random_seed: object

seed for random number generator

kwargs: dict

additional parameters for the sampling method

property method_name: str

Name of sampling method.

sampling()

Creates a new instance of the respective sampling method, and calls its respective extract_subgraph method.

class netin.sampling.RandomEdges(g: Graph, pseeds: float, max_tries: int = 20, random_seed: object | None = None, **kwargs)

Random edge sampling.

Parameters

g: netin.Graph | netin.DiGraph

global network

pseeds: float

fraction of seeds to sample

max_tries: int

maximum number of tries to sample a subgraph with enough classes and edges

random_seed: object

seed for random number generator

kwargs: dict

additional parameters for the sampling method

property method_name: str

Name of sampling method.

sampling()

Creates a new instance of the respective sampling method, and calls its respective extract_subgraph method.

class netin.sampling.RandomNeighbor(g: Graph, pseeds: float, max_tries: int = 20, random_seed: object | None = None)

Random neighbor sampling.

Parameters

g: netin.Graph | netin.DiGraph

global network

pseeds: float

fraction of seeds to sample

max_tries: int

maximum number of tries to sample a subgraph with enough classes and edges

random_seed: object

seed for random number generator

kwargs: dict

additional parameters for the sampling method

property method_name: str

Name of sampling method.

sampling()

Creates a new instance of the respective sampling method, and calls its respective extract_subgraph method.

class netin.sampling.DegreeRank(g: Graph, pseeds: float, random_seed: object | None = None, **kwargs)

Sampling by degree rank

Parameters

g: netin.Graph | netin.DiGraph

global network

pseeds: float

fraction of seeds to sample

max_tries: int

maximum number of tries to sample a subgraph with enough classes and edges

random_seed: object

seed for random number generator

kwargs: dict

additional parameters for the sampling method

order: str

order of nodes by degree. Options: “asc” | “desc”

property method_name: str

Name of sampling method.

sampling()

Creates a new instance of the respective sampling method, and calls its respective extract_subgraph method.

class netin.sampling.DegreeGroupRank(g: Graph, pseeds: float, random_seed: object | None = None, **kwargs)

Sampling by degree group rank (by degree rank per class)

Parameters

g: netin.Graph | netin.DiGraph

global network

pseeds: float

fraction of seeds to sample

max_tries: int

maximum number of tries to sample a subgraph with enough classes and edges

random_seed: object

seed for random number generator

kwargs: dict

additional parameters for the sampling method

order: str

order of nodes by degree. Options: “asc” | “desc”

property method_name: str

Name of sampling method.

sampling()

Creates a new instance of the respective sampling method, and calls its respective extract_subgraph method.

class netin.sampling.PartialCrawls(g: Graph, pseeds: float, max_tries: int = 20, random_seed: object | None = None, **kwargs)

Sampling by partial crawls.

Parameters

g: netin.Graph | netin.DiGraph

global network

pseeds: float

fraction of seeds to sample

max_tries: int

maximum number of tries to sample a subgraph with enough classes and edges

random_seed: object

seed for random number generator

kwargs: dict

additional parameters for the sampling method

References

[Yang2017]

10. Yang, B. Ribeiro, & J. Neville “Should We Be Confident in Peer Effects Estimated From Social Network Crawls?” ICWSM (Vol. 11, No. 1, pp. 708-711), 2017.

property method_name: str

Name of sampling method.

sampling()

Creates a new instance of the respective sampling method, and calls its respective extract_subgraph method.