In [11]:
from bigraph_viz import plot_bigraph
plot_settings = {}
example¶
This example uses a composite specification that has nesting, multiple processes, types and values included.
In [12]:
nested_composite_spec = {
'store1': {
'store1.1': {
'_value': 1.1,
'_type': 'float',
},
'store1.2': {
'_value': 2,
'_type': 'integer',
},
'store1.3': {
'_value': 1.1,
'_type': 'float',
},
'process1': {
'_type': 'process',
'_outputs': {
'port1': 'float',
'port2': 'integer',
},
'outputs': {
'port1': 'store1.1',
'port2': 'store1.2',
}
},
'process2': {
'_type': 'process',
'_outputs': {
'port1': 'float',
'port2': 'integer',
},
'outputs': {
'port1': 'store1.1',
'port2': 'store1.2',
}
},
},
'process3': {
'_type': 'process',
'outputs': {
'port1': 'store1',
}
}
}
plot_bigraph(nested_composite_spec)
Out[12]:
Processes are by default shown with place edges to their outer nodes (see process1 and process2 above). To remove this use remove_process_place_edges=True:
In [13]:
plot_settings['remove_process_place_edges'] = True
plot_bigraph(nested_composite_spec, **plot_settings)
Out[13]:
To remove specific nodes from the vizualization, pass in a list of paths to those nodes. In this example the node in path ('store1', 'store1.3') is removed:
In [14]:
remove_nodes = [
('store1', 'store1.3'),
]
plot_settings['remove_nodes'] = remove_nodes
plot_bigraph(nested_composite_spec, **plot_settings)
Out[14]:
Remove port labels:
In [15]:
plot_settings['port_labels'] = False
plot_bigraph(nested_composite_spec, **plot_settings)
Out[15]:
Change node borders or fill colors
In [16]:
plot_settings['node_border_colors'] = {
('store1', 'store1.1'): 'blue',
}
plot_settings['node_fill_colors'] = {
('store1', 'store1.2'): 'red',
}
plot_bigraph(nested_composite_spec, **plot_settings)
Out[16]:
To group nodes use node_groups, A list of lists of grouped node paths, which will be visually placed at the same rank.
In [17]:
plot_settings['node_groups'] = [
[('store1', 'process1'), ('store1', 'process2'), ('process3',)]
]
plot_bigraph(nested_composite_spec, **plot_settings)
Out[17]:
Change the orientation with rankdir using the graphviz orientation key. 'TB'=top-to-bottom, 'LR'=left-to-right. By default this is top-to-bottom.
In [18]:
plot_settings['rankdir'] = 'LR'
plot_bigraph(nested_composite_spec, **plot_settings)
Out[18]:
In [ ]: