aboutsummaryrefslogtreecommitdiffhomepage
path: root/tensorflow/contrib/rnn
diff options
context:
space:
mode:
authorGravatar Billy Lamberta <blamb@google.com>2018-09-06 15:55:13 -0700
committerGravatar TensorFlower Gardener <gardener@tensorflow.org>2018-09-06 16:03:41 -0700
commit75bc3006b890bfc9c58a05097a7bce10bb30c17e (patch)
tree70863cf1a48f70d70cd3a7643b9a31fc95bf7be0 /tensorflow/contrib/rnn
parent1e9cbcc208a85c467b3db7fbbcba681aa012c607 (diff)
Update LSTM paper reference
Match #22072 PiperOrigin-RevId: 211884527
Diffstat (limited to 'tensorflow/contrib/rnn')
-rw-r--r--tensorflow/contrib/rnn/python/ops/rnn_cell.py20
1 files changed, 11 insertions, 9 deletions
diff --git a/tensorflow/contrib/rnn/python/ops/rnn_cell.py b/tensorflow/contrib/rnn/python/ops/rnn_cell.py
index f74c95f962..06c481672c 100644
--- a/tensorflow/contrib/rnn/python/ops/rnn_cell.py
+++ b/tensorflow/contrib/rnn/python/ops/rnn_cell.py
@@ -97,10 +97,10 @@ class CoupledInputForgetGateLSTMCell(rnn_cell_impl.RNNCell):
The default non-peephole implementation is based on:
- http://www.bioinf.jku.at/publications/older/2604.pdf
+ https://pdfs.semanticscholar.org/1154/0131eae85b2e11d53df7f1360eeb6476e7f4.pdf
- S. Hochreiter and J. Schmidhuber.
- "Long Short-Term Memory". Neural Computation, 9(8):1735-1780, 1997.
+ Felix Gers, Jurgen Schmidhuber, and Fred Cummins.
+ "Learning to forget: Continual prediction with LSTM." IET, 850-855, 1999.
The peephole implementation is based on:
@@ -2448,10 +2448,10 @@ class LayerNormLSTMCell(rnn_cell_impl.RNNCell):
The default non-peephole implementation is based on:
- http://www.bioinf.jku.at/publications/older/2604.pdf
+ https://pdfs.semanticscholar.org/1154/0131eae85b2e11d53df7f1360eeb6476e7f4.pdf
- S. Hochreiter and J. Schmidhuber.
- "Long Short-Term Memory". Neural Computation, 9(8):1735-1780, 1997.
+ Felix Gers, Jurgen Schmidhuber, and Fred Cummins.
+ "Learning to forget: Continual prediction with LSTM." IET, 850-855, 1999.
The peephole implementation is based on:
@@ -2802,9 +2802,11 @@ class WeightNormLSTMCell(rnn_cell_impl.RNNCell):
Training of Deep Neural Networks
The default LSTM implementation based on:
- http://www.bioinf.jku.at/publications/older/2604.pdf
- S. Hochreiter and J. Schmidhuber.
- "Long Short-Term Memory". Neural Computation, 9(8):1735-1780, 1997.
+
+ https://pdfs.semanticscholar.org/1154/0131eae85b2e11d53df7f1360eeb6476e7f4.pdf
+
+ Felix Gers, Jurgen Schmidhuber, and Fred Cummins.
+ "Learning to forget: Continual prediction with LSTM." IET, 850-855, 1999.
The class uses optional peephole connections, optional cell clipping
and an optional projection layer.