kunwuz
commented
9 months ago Hi, thanks for the question. The output of GES is a Markov Equivalence Class, which means that there may exist edges of which the directions cannot be determined by the algorithm. It seems that the algorithm cannot determine any direction for your data.
If you would like to get direction for every edge, perhaps you may consider functional-constraint-based methods, such as LiNGAM. A usage case of LiNGAM can be found in this notebook.
MattFill
jdramsey
I have been experimenting with the GES algorithm to estimate causal paths from cross-sectional data. The following code produces the graph attached. I am wondering how I can estimate the causal direction between nodes from GES and visualize this?
Set up
from causallearn.search.ScoreBased.GES import ges
data = df_dx[['Bio','Psycho','Dx','NumberChronicPainTypes_T0']]
default parameters
Record = ges(data)
or customized parameters
scoring = 'local_score_BIC' maxP = None # maximum number of parents when searching the graph parameters = None Record = ges(data, scoring, maxP, parameters)
Visualization using pydot
from causallearn.utils.GraphUtils import GraphUtils import matplotlib.image as mpimg import matplotlib.pyplot as plt import io
pyd = GraphUtils.to_pydot(Record['G'],labels=['Bio','Psycho','Dx','NumberChronicPainTypes_T0']) tmp_png = pyd.create_png(f="png") fp = io.BytesIO(tmp_png) img = mpimg.imread(fp, format='png') plt.axis('off') plt.imshow(img) plt.show()
or save the graph
pyd.write_png('simple_test.png')