You can subtract the outcome of transform() from 1 to flip the labels of your data. Check the following example.
>>> from sklearn import preprocessing
>>> le = preprocessing.LabelEncoder()
>>> Y=['p','p','e','e','e','p','e','p']
>>> le.fit(Y)
LabelEncoder()
>>> le.transform(Y) #get numeric label
array([1, 1, 0, 0, 0, 1, 0, 1], dtype=int64)
>>> 1-le.transform(Y) # flip the label
array([0, 0, 1, 1, 1, 0, 1, 0], dtype=int64)
>>>