Usage
The toolbox is designed in a modular way, which means the individual modules can be used in combination with others, but also by themselves. Users also have the option to choose between a normal Python scripting API and a graphical user interface (GUI).
Tip
The GUI provides convenient tools for data handling, (dynamic) connectivity estimation, graph analysis, and multiverse exploration, offering a quick visual overview. For greater flexibility, the scripting API is recommended.
GUI
After installation, graphical user interface can be accessed through the terminal by typing:
comet-gui
On a fresh installation, the first startup may take a little longer due to the initialisation of the Matplotlib backend. A brief introduction for how to use the GUI is provided in the Tutorials.
Scripting
Dynamic functional connectivity can be estimated through the connectivity module. An example for sliding window correlation:
from comet import connectivity, utils
ts = utils.load_example()
sw = connectivity.SlidingWindow(ts, windowsize=30, shape="gaussian")
dfc = sw.estimate()
Graph measures can be calculated through the graph module. An example for the clustering coefficient derived from sliding window estimates:
from comet import connectivity, graph, utils
ts = utils.load_example()
sw = connectivity.SlidingWindow(ts, windowsize=30, shape="gaussian")
dFC = sw.estimate()
adj = graph.threshold(dFC, type="density", threshold=0.2)
clustering_coef = graph.clustering_coef(adj)
Multiverse analysis can be conducted through the multiverse module. This exaple will create and run a multiverse analysis with two decisions (6 possible combinations):
from comet.multiverse import Multiverse
forking_paths = {
"decision1": [1, 2, 3],
"decision2": ["Hello", "World"]
}
def analysis_template():
print("Decision1:", {{decision1}})
print("Decision2", {{decision2}})
mverse = Multiverse(name="example_multiverse")
mverse.create(analysis_template, forking_paths)
mverse.summary()
mverse.run()