API

SignLens

class sign_lens.sign_lens.SignLens(edgelist_fpath, seperator='\t', header=None)

SignLens is a class for analyzing signed networks.

calc_balanced_triangle_dist() → tuple

calculate balanced triangle distributions

Returns:

tuple

(balanced triads, unbalanced triads)

calc_edge_num() → int

calculate the number of edges

Returns:

int

the edge number

calc_hop_dist() → dict

calculate the distrubiton of hops

Returns:

dict

the dict of {'d': counts }

calc_node_num() → int

calculate the number of nodes

Returns:

int

the node number

calc_sign_dist() → tuple

calculate sign distribution

Returns:

tuple

(positive edge number, negative edge number, pos_neg_ratio)

calc_signed_in_degree() → tuple

calculate signed in degree

Returns:

tuple

(G_in_degree, pos_G_in_degree, neg_G_in_dergee)

calc_signed_out_degree() → tuple

calculate signed out degree

Returns:

tuple

(G_in_degree, pos_G_in_degree, neg_G_in_dergee)

calc_signed_triads_dist() → tuple

calculate signed triads distributions

Returns:

tuple

((+++, ++-, +–, —), balanced triads, unbalanced triads)

calc_singular_value_dist() → array

calculated singular value distribution

Returns:

return the svd results of undirected unsigned matrice

report_signed_metrics(output_dir='output') → str

Report signed metrics for a signed network.

The main signed network metrics include sign distribution, balanced triangle distrubition, signed in-degree distribution, signed out-degree distribution, in-degree distribution, out-degree distribution, hop plot and singular value distribution according to this paper: “BalanSiNG: Fast and Scalable Generation of Realistic Signed Networks”.

Parameters:

output_dirstr, optional

It will output some figures to the ourput_dir, by default ‘output’

Returns:

str

The table for signed metrics.

SignedTriadFeaExtra

class sign_lens.utils.SignedTriadFeaExtra(edgelist_fpath, undirected=False, seperator='\t')

extract_triad_counts(u, v) → tuple

\[A \times B \alpha\]