tfa.text.crf_log_likelihood

Computes the log-likelihood of tag sequences in a CRF.

View aliases

Main aliases

tfa.text.crf.crf_log_likelihood

tfa.text.crf_log_likelihood(
    inputs: tfa.types.TensorLike,
    tag_indices: tfa.types.TensorLike,
    sequence_lengths: tfa.types.TensorLike,
    transition_params: Optional[TensorLike] = None
) -> Tuple[tf.Tensor, tf.Tensor]

Args
inputs	A [batch_size, max_seq_len, num_tags] tensor of unary potentials to use as input to the CRF layer.
tag_indices	A [batch_size, max_seq_len] matrix of tag indices for which we compute the log-likelihood.
sequence_lengths	A [batch_size] vector of true sequence lengths.
transition_params	A [num_tags, num_tags] transition matrix, if available.

Returns
log_likelihood	A [batch_size] Tensor containing the log-likelihood of each example, given the sequence of tag indices.
transition_params	A [num_tags, num_tags] transition matrix. This is either provided by the caller or created in this function.