SHAP: How to get feature importance for categorical variable?

I am trying to use the SHAP for an ANN model interpretation. I found that when I using

shap.summary_plot(shap_values[0], backgournd, plot_type='bar')

For a categorical variable after dummy, the feature importance would be something like

CategoricalFeature_A=0.1
CategoricalFeature_B=0.3
CategoricalFeature_C=0.3

I am wondering is there any way I can directly get the feature importance of CategoricalFeature ?