crf_multitag_sequence_score: CRF multitag sequence score

In tfaddons: Interface to 'TensorFlow SIG Addons'

Description

Computes the unnormalized score of all tag sequences matching

Usage

crf_multitag_sequence_score(
  inputs,
  tag_bitmap,
  sequence_lengths,
  transition_params
)

Arguments

inputs	A [batch_size, max_seq_len, num_tags] tensor of unary potentials to use as input to the CRF layer.
tag_bitmap	A [batch_size, max_seq_len, num_tags] boolean tensor representing all active tags at each index for which to calculate the unnormalized score.
sequence_lengths	A [batch_size] vector of true sequence lengths.
transition_params	A [num_tags, num_tags] transition matrix.

Details

tag_bitmap. tag_bitmap enables more than one tag to be considered correct at each time step. This is useful when an observed output at a given time step is consistent with more than one tag, and thus the log likelihood of that observation must take into account all possible consistent tags. Using one-hot vectors in tag_bitmap gives results identical to crf_sequence_score.

Value

sequence_scores: A [batch_size] vector of unnormalized sequence scores.

tfaddons documentation built on July 2, 2020, 2:12 a.m.