projects / platform / upstream / tensorflow.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 5e9c94c) | patch
Use sequence_length arg for dynamic_rnn within RNNEstimator
authorA. Unique TensorFlower <gardener@tensorflow.org>
Wed, 16 May 2018 19:24:59 +0000 (12:24 -0700)
committerTensorFlower Gardener <gardener@tensorflow.org>
Wed, 16 May 2018 19:27:53 +0000 (12:27 -0700)
commitd825bd42b2fce3a139c05e1412ab6a6b334ff52c
treec5f015a17a9d9eb3952353b5ae12ed904e78d7edtree | snapshot
parent5e9c94cb7c181f4564d6718cab7a309453ecafe9commit | diff
Use sequence_length arg for dynamic_rnn within RNNEstimator

This does not affect correctness, but improves performance by skipping padded
parts of the sequence. Also correct documentation in rnn.py that states the
opposite.

PiperOrigin-RevId: 196869793
tensorflow/contrib/estimator/python/estimator/rnn.py diff | blob | history
tensorflow/python/ops/rnn.py diff | blob | history
Domain: Machine Learning / ML Framework;