irisclustering.py 文件源码

def GetFeatures(frame):
    #convert data to float
    arr = np.array(frame,dtype=np.float)
    #fill missing values
    from sklearn.preprocessing import Imputer
    imputer = Imputer(strategy='mean')
    arr = imputer.fit_transform(arr)
    #normalize the entire data
    from sklearn.preprocessing import scale
    arr = scale(arr)
    return arr
#=================================================