Predict sequence when some parts are known

[dimidd]

Is your feature request related to a problem? Please describe.
I'd like to predict labels for a sequence. Suppose some of the labels are known, and I only want to predict the rest. E.g. in the sentece
'Paxar Corp said it has acquired Thermo-Print GmbH'
Suppose I know that 'Paxar Corp' is a named entity of type ORG, but I'd still like to predict other entities. Assuming that the knowledge of some entities can affect the label distribution of others.

Describe the solution you'd like
Perhaps add an optional param named known to SequenceLabeler.predict, which is a list of dictionaries. Each dictionary will contain the start and end indices, the text, and the label.