What would I need to do to change the colors of the classes? Maybe provide a simple ex

improve documentation on coloring about dtreeviz HOT 7 CLOSED

parrt commented on September 13, 2024

improve documentation on coloring

from dtreeviz.

Comments (7)

DJCordhose commented on September 13, 2024 1

Very good, I can now do what I want! Thank you so much!

from dtreeviz.

Ashafix commented on September 13, 2024

Did you have a look at the end of the readme file and notebooks/colors.ipynb?

If it doesn't answer your question, it should be added to the documentation.

from dtreeviz.

DJCordhose commented on September 13, 2024

I have and have even read source code and made some experiments, but did not get it to work. Also looked at a notebook that did some coloring, but simply did not understand it.

from dtreeviz.

parrt commented on September 13, 2024

Sounds like we need some simpler examples.

from dtreeviz.

DJCordhose commented on September 13, 2024

Or maybe the API is just too complex. I don't know, because I still don't understand 😀

from dtreeviz.

Ashafix commented on September 13, 2024

Right now the documentation says about the class colors:

classes needs to be a list of lists of colors with a minimum length of your number of colors. The index is the number of classes and the list with this index needs to have the same amount of colors.

The usage would be

class_colors = [None, # 0 classes
                None, # 1 class
                ["#FEFEBB","#a1dab4"], # 2 classes
                ["#ff0000","#00ff00",'#0000ff'], # colors were changed here
                ["#FEFEBB","#D9E6F5",'#a1dab4','#fee090'], # 4
                ["#FEFEBB","#D9E6F5",'#a1dab4','#41b6c4','#fee090'], # 5
                ["#FEFEBB",'#c7e9b4','#41b6c4','#2c7fb8','#fee090','#f46d43'], # 6
                ["#FEFEBB",'#c7e9b4','#7fcdbb','#41b6c4','#225ea8','#fdae61','#f46d43'], # 7
                ["#FEFEBB",'#edf8b1','#c7e9b4','#7fcdbb','#1d91c0','#225ea8','#fdae61','#f46d43'], # 8                                       ["#FEFEBB",'#c7e9b4','#41b6c4','#74add1','#4575b4','#313695','#fee090','#fdae61','#f46d43'], # 9
                ["#FEFEBB",'#c7e9b4','#41b6c4','#74add1','#4575b4','#313695','#fee090','#fdae61','#f46d43','#d73027'] # 10
]


classifier = tree.DecisionTreeClassifier(max_depth=2)
iris = load_iris()
classifier.fit(iris.data, iris.target)

dtreeviz.trees.dtreeviz(classifier,
                        iris.data,
                        iris.target,
                        target_name='variety',
                        feature_names=iris.feature_names,
                        class_names=["setosa", "versicolor", "virginica"],
                        colors={'classes': class_colors}
)

Any ideas on how to improve the documentation are appreciated and I'd be happy to add it to the documentation.

from dtreeviz.

DJCordhose commented on September 13, 2024

What about simplifying that example to something much more basic as a first step?

from dtreeviz.

Recommend Projects