tfa.text.viterbi_decode

Decode the highest scoring sequence of tags outside of TensorFlow.

tfa.text.viterbi_decode(
    score: tfa.types.TensorLike,
    transition_params: tfa.types.TensorLike
) -> tf.Tensor

This should only be used at test time.

score A [seq_len, num_tags] matrix of unary potentials.
transition_params A [num_tags, num_tags] matrix of binary potentials.

viterbi A [seq_len] list of integers containing the highest scoring tag indices.
viterbi_score A float containing the score for the Viterbi sequence.