python - How to use sci-kit Decision trees for unsupervised classification -

I want to use a decision tree to create a group of Excel spreadsheets to group groups into groups like file Name of sheet, number sheet, sheet 1

I want to use the science-decision decision tree classifier Each sample I supply is an example of a sample of my data data list here. 1}

I have about 300 samples to classify and I want to use this decision DecisionTreeClassifier. Because the data contains the strings I

  vec = DictVectorizer () train = vec.fit_transform (decisionData) .toarray ()  

I also want to use To split the samples, start with the entropy criterion and the maximum of _ 3 of the path

  clf = decisionTreeClassifier (criteria = 'entropy', max_depth = 3)  

Now I need to call

  clf = clf.fit (train, label)  

I do not know what type of label vector Should take in. Is this an ID only for each sample?

In addition, when I get the result, how do I remove the leaf on the tree for each sample which has been supplied. I am worried about how the label should be transformed so that I can understand what the classification was.

I can see that all documents are for supervision, where a known goal has been given so that any help can be great.