Can use either greedy search (also known as best path) or a constrained dictionary search.
k_ctc_decode( y_pred, input_length, greedy = TRUE, beam_width = 100L, top_paths = 1 )
(samples, time_steps, num_categories) containing the
prediction, or output of the softmax.
(samples, ) containing the sequence length for
each batch item in
perform much faster best-path search if
TRUE. This does not
use a dictionary.
FALSE: a beam search decoder will be used
with a beam of this width.
FALSE, how many of the most probable paths
will be returned.
TRUE, returns a list of one element
that contains the decoded sequence. If
FALSE, returns the
top_pathsmost probable decoded sequences. Important: blank labels are returned as
(top_paths) that contains the log probability of each
This function is part of a set of Keras backend functions that enable lower level access to the core operations of the backend tensor engine (e.g. TensorFlow, CNTK, Theano, etc.).
You can see a list of all available backend functions here: https://keras.rstudio.com/articles/backend.html#backend-functions.