CODEWITHSUNDEEP
pandasseaborn
mrvaita
mrvaita

Reputation: 103

Seaborn clustermap doesn't show row colors

I generated a clustermap using seaborn.clustermap. I wanted to show the row_colors option to highlight some clusters but this is the result I got:

clustermap with missing row_colors

Here you can find my code:

pal = sns.light_palette('red', np.unique(labels).size)
lut = dict(zip(np.unique(labels), pal))
row_colors = pd.Series(labels, name='clusters').map(lut)
sns.clustermap(my_data, method='ward', robust=True, row_colors=row_colors)

However if I run the example from the seaborn documentation I don't have any problem:

enter image description here

iris = sns.load_dataset("iris")
species = iris.pop("species")
lut = dict(zip(species.unique(), "rbg"))
row_colors = species.map(lut)
g = sns.clustermap(iris, row_colors=row_colors)

Why does the highlighting not work in my code?

Upvotes: 2

Views: 3345

Answers (1)

mrvaita
mrvaita

Reputation: 103

Thank you very much for you answer error. I found out the problem. I have a MultiIndex dataframe and for some reason it doesn't plot the row_color. Actually this is the only difference with the iris example code.

And I just fixed the problem doing this:

sns.clustermap(my_data.reset_index(drop=True), method='ward', robust=True, row_colors=row_colors)

and now it works:

enter image description here

I don't know whether this can be considered as a bug but it looks like. Perhaps that can help to fix it.

Upvotes: 6

Related Questions