blugraph.core.io module

Reference page for the bluegraph.core.io module.

Collection of data structures for representing property graphs as data frames.

class bluegraph.core.io.GraphProcessor(pgframe=None, directed=True)

Abstract class for a graph processor.

The provided interface allows to convert PGFrames into backend-specific graph objects and vice versa. It also allows to access nodes/edges and properties of backend-specific graph objects through its interface.

exception ProcessorException
classmethod from_graph_object(graph_object)

Initialize directly from the input graph object.

get_pgframe(node_prop_types=None, edge_prop_types=None, node_filter=None, edge_filter=None)

Get a new pgframe object from the wrapped graph object.

abstract neighbors(node_id)

Get neighors of the node.

class bluegraph.core.io.PGFrame(nodes=None, edges=None, default_prop_type='category')

Class for storing typed property graphs as a collection of frames.

exception PGFrameException
abstract copy()

Copy the PGFrame.

abstract edge_properties(edge_type=None, include_type=None)

Return a list of edge properties.

abstract edge_types(flatten=False)

Return a set of edges types.

abstract edges(typed_by=None, raw_frame=False)

Return a list of edges.

export_to_gephi(prefix, node_attr_mapping, edge_attr_mapping, edge_filter=None)

Save the graph for Gephi import.

Saves the graph as two .csv files one with nodes (<prefix>_nodes.csv) and one with edges ( <prefix>_edges.csv). Node IDs are replaced by interger identifiers (Gephi asks for node IDs to be numerical) and entity names are added as the node property ‘Label’.

from_jsonld(resources, include_context=True, type_handler=None, types_from_relations=True, exclude=None, only_props=False)

Create a PGFrame from jsonld.

Parameters
  • resources (iterable of dict) – Collection of input resources in JSON-LD format

  • include_context (bool, optional) – Flag indicating if the context should be included as a property. Default is True.

  • type_handler (func, optional) – Function to apply to the value of type (e.g. @type’)

  • types_from_relations (bool, optional) – Flag indicating if resources with unkown types should be assigned with types from the incoming relations. Default is True

  • exclude (list of str, optional) – Collection of property names to exclude. Default is empty.

  • only_props (bool, optional) – Flag indicating if the procedure should extract only properties from the given resources.

classmethod from_ontology(filepath=None, rdf_graph=None, format='turtle', remove_prop_uris=False)

Create a PandasPGFrame from ontology.

abstract node_properties(node_type=None, include_type=None)

Return a list of node properties

abstract node_types(flatten=False)

Return a set of node types.

abstract nodes(typed_by=None, raw_frame=False)

Return a list of nodes.

abstract number_of_edges(edge_type=None)

Return a number of nodes.

abstract number_of_nodes(node_type=None)

Return a number of nodes.

abstract rename_edge_properties(mapping)

Rename edge properties.

abstract rename_node_properties(mapping)

Rename node properties.

to_jsonld(edges_key='edges')

Create a JSON-LD representation of the PGFrame.

class bluegraph.core.io.PandasPGFrame(nodes=None, edges=None, default_prop_type='category')

Class for storing typed PGs as a collection of pandas DataFrames.

add_edges(edges)

Add edge ids to the PG frame.

add_edges_from_df(df, source_column, target_column, edge_type=None, property_types=None)

Add edges from a dataframe.

add_nodes(node_ids)

Add node ids to the PG frame.

add_nodes_from_df(df, id_column, node_type=None, property_types=None)

Add nodes from a dataframe.

copy()

Create a copy of the pgframe.

edge_properties(edge_type=None, include_type=False)

Return a list of edge properties.

edge_types(flatten=False)

Return a list of edges types.

edges(typed_by=None, raw_frame=False, include_index=False, filter_props=None, rename_cols=None)

Return a list of edges.

export_json(path)

Export to a JSON file.

filter_edges(edges)

Get edges included in the input list.

filter_nodes(nodes)

Get nodes included in the input list.

classmethod from_frames(nodes, edges, node_prop_types=None, edge_prop_types=None)

Initialize from node/edge dataframes.

classmethod from_json(json_data)

Load from a JSON dictionary.

get_edge(s, t)

Get edge properties.

get_node(n)

Get node properties.

has_edge_types()

Return flag indicating if edges are typed.

has_node_types()

Return flag indicating if nodes are typed.

isolated_nodes()

Get nodes without any incident edges.

classmethod load_json(path)

Load from a JSON file.

node_properties(node_type=None, include_type=False)

Return a list of node properties

node_types(flatten=False)

Return a list of node types.

nodes(typed_by=None, raw_frame=False, include_index=False, filter_props=None, rename_cols=None)

Return a list of nodes.

number_of_edges()

Return a number of nodes.

number_of_nodes()

Return a number of nodes.

remove_edges(edges_to_remove)

Remove edges from the graph.

remove_isolated_nodes()

Remove nodes without any incident edges.

remove_nodes(nodes_to_remove)

Remove nodes from the graph.

rename_edge_properties(mapping)

Rename edge properties.

rename_node_properties(mapping)

Rename node properties.

subgraph(nodes=None, edges=None, remove_isolated_nodes=False)

Get a subgraph induced by input nodes and edges.

to_json(node_id_key=None, node_type_key=None, edge_id_keys=None, edge_type_key=None)

Covert to a JSON dictionary.

to_triples(predicate_prop='@type', include_type=True, include_literals=True)

Generate triples from the PGframe.

class bluegraph.core.io.SparkPGFrame

Class for storing typed PGs as a collection of Spark DataFrames.