diff options
author | Eugene Brevdo <ebrevdo@google.com> | 2017-03-30 11:35:17 -0800 |
---|---|---|
committer | TensorFlower Gardener <gardener@tensorflow.org> | 2017-03-30 12:47:55 -0700 |
commit | 052e207e4f1076ac6b7464c01a67ecf4f986766d (patch) | |
tree | d84e0d28d9e1f8a424458ac41b284779b78627f9 /third_party/snappy.BUILD | |
parent | d82f8f818cec7a1099a7e680a30ba8f5f8ed2589 (diff) |
Revert DynamicRNNEstimator's ability to accept an instance of RNNCell.
As we move to make RNNCells subclasses of Layers, several things become clear:
1. Layers are stateful - once called, they keep track of their variables.
2. Estimators create a new graph for predict/fit/etc.
3. Items #1 and #2 are not compatible, because a Layer modified in .fit()
cannot be used in .predict(): its variables and internal state tensors
are associated with the graph created in fit() -- and cannot be used
with the graph being created in predict(). Strange errors occur.
As a result, Estimators may never accept a Layer instance given this API
conflict.
This PR unblocks the move of RNNCell to be subclasses of tf.layers.Layer.
Change: 151735868
Diffstat (limited to 'third_party/snappy.BUILD')
0 files changed, 0 insertions, 0 deletions