The function stacks the SHAP values for each observation and shows how the final output was obtained as a sum of each predictor’s attributions through the force plot. (Randomly plotting a certain portion of the data is optional in case the dataset is large.)
Arguments
- shap_se
A SummarizedExperiment object with results computed by
ml_shap
.- top_feature
Integer. Top number of features to be shown. If top_feature is set to greater than 10, only the top 10 samples will be displayed. Default is
10
.- cluster_method
Character. The clustering method. Allowed methods include "ward.D", "ward.D2", "single", "complete", "average", "mcquitty", "median", "centroid". Default is
'ward.D'
.- group_num
Integer. The numbers of groups to be shown in the plots. Default is
10
.