Reputation: 41
I have used Decision Tree to solve a problem. Then I have used graphviz to obtain the pictorial version of the decision tree.
import graphviz
dot_data = tree.export_graphviz(clf, out_file=None)
graph = graphviz.Source(dot_data)
dot_data = tree.export_graphviz(clf, out_file=None,
feature_names=f_name,
class_names=['Topper', 'Not a topper'],
filled=True, rounded=True,
special_characters=True)
graph = graphviz.Source(dot_data)
graph
I wish to download this tree generated in .png, .jpg, or any suitable format. Is it possible?
Upvotes: 4
Views: 10411
Reputation: 1
dot.render('/mnt/data/fluxograma_sac_logistico', format='png', cleanup=False) '/mnt/data/fluxograma_sac_logistico.png'
Upvotes: -1
Reputation: 187
From digraph source create an image
digraph source -
graph_data = "digraph G { ...... }"
fie_ext = 'png'
temp_img = 'temp_file'
my_graph = graphviz.Source(graph_data)
my_graph.render(temp_img,format=fie_ext, view=False)
render() method auto generate an image with temp_file.png extension in your current directory.
Upvotes: 5
Reputation: 595
To export dot file to image, you need module pydotplus.
from sklearn.tree import export_graphviz
from sklearn.externals.six import StringIO
import pydotplus
dot_data = StringIO()
export_graphviz(clf, out_file=dot_data,
filled=True, rounded=True,
special_characters=True,feature_names = feature_cols,class_names=['0','1'])
graph = pydotplus.graph_from_dot_data(dot_data.getvalue())
graph.write_png('my_decision_tree.png')
Upvotes: 3