DYMOND.learning package¶
Submodules¶
DYMOND.learning.learn_parameters module¶
- DYMOND.learning.learn_parameters.calc_motifs_timesteps(motif_types, g, timesteps, tmp_files_dir)[source]¶
Calculate timesteps each motif appears in
- Parameters
motif_types (dict) – motif types
g (igraph.Graph) – input graph
timesteps (int) – graph timesteps
tmp_files_dir (str) – directory for tmp files
- DYMOND.learning.learn_parameters.get_active_nodes(g, timesteps, tmp_files_dir)[source]¶
Get active nodes in each timestep
- Parameters
g (igraph.Graph) – input graph
timesteps (int) – graph timesteps
tmp_files_dir (str) – directory for tmp files
- Returns
active nodes per timestep
- Return type
dict
- DYMOND.learning.learn_parameters.get_dataset(dataset_dir)[source]¶
Get dataset directory, info, and graph.
- Parameters
dataset_dir (str) – dataset directory
- Returns
dataset directory, info, and graph
- Return type
str, dict, igraph.Graph
- DYMOND.learning.learn_parameters.get_directories_parameters(dataset_dir)[source]¶
Get the directories for parameters and temp save files. If they don’t exist, create the directories.
- Parameters
dataset_dir (str) – dataset directory
- Returns
parameters directory, tmp files directory
- Return type
str, str
- DYMOND.learning.learn_parameters.get_motif_counts(motifs, motif_types, g, timesteps, tmp_files_dir)[source]¶
Estimate motif edge-weighted counts
- Parameters
motifs (list) – motifs
motif_types (dict) – motif types
g (igraph.Graph) – input graph
timesteps (int) – graph timesteps
tmp_files_dir (str) – directory for tmp files
- Returns
motif edge-weighted counts and num. of timesteps
- DYMOND.learning.learn_parameters.get_motifs_graph(g, timesteps, tmp_files_dir, nodes=None)[source]¶
Get the motifs in the input graph
- Parameters
g (igraph.Graph) – input graph
timesteps (int) – graph timesteps
tmp_files_dir (str) – directory for tmp files
nodes (list) – (optional) nodes to get motifs for
- Returns
motifs and motif types
- Return type
dict
- DYMOND.learning.learn_parameters.get_motifs_t(t, tmp_files_dir)[source]¶
Get motifs in timestep t
- Parameters
t (int) – timestep
tmp_files_dir (str) – directory for tmp files
- Returns
motifs at time t
- Return type
dict
- DYMOND.learning.learn_parameters.get_node_role_counts(g, timesteps, tmp_files_dir)[source]¶
Estimate node role counts
- Parameters
g (igraph.Graph) – input graph
timesteps (int) – graph timesteps
tmp_files_dir (str) – directory for tmp files
- Returns
node role counts, motif type counts
- Return type
dict, dict
- DYMOND.learning.learn_parameters.learn_motif_interarrival_rates(motifs, motif_types, g, timesteps, tmp_files_dir)[source]¶
Estimate inter-arrival rates per motif type
- Parameters
motifs (list) – motifs
motif_types (dict) – motif types
g (igraph.Graph) – input graph
timesteps (int) – graph timesteps
tmp_files_dir (str) – directory for tmp files
- Returns
motif interarrival rates
- Return type
dict
- DYMOND.learning.learn_parameters.learn_motif_proportions(motifs, motif_types, size_V, tmp_files_dir)[source]¶
Estimate proportions of each motif type
- Parameters
motifs (list) – motifs
motif_types (dict) – motif types
size_V (int) – number of nodes
tmp_files_dir –
- Returns
motif type proportions
- Return type
dict
- DYMOND.learning.learn_parameters.learn_node_arrival_rates(g, timesteps, tmp_files_dir)[source]¶
Estimate node arrival rates
- Parameters
g (igraph.Graph) – input graph
timesteps (int) – graph timesteps
tmp_files_dir (str) – directory for tmp files
- Returns
node arrival rate
- Return type
dict
- DYMOND.learning.learn_parameters.learn_node_roles_distribution(g, timesteps, tmp_files_dir)[source]¶
Estimate node role probabilities
- Parameters
g (igraph.Graph) – input graph
timesteps (int) – graph timesteps
tmp_files_dir (str) – directory for tmp files
- Returns
node role probabilities, role counts, motif type counts
- Return type
dict, dict, dict