diff options
| author |  A. Unique TensorFlower <gardener@tensorflow.org> | 2017-11-27 04:26:18 -0800 |
|---|---|---|
| committer | TensorFlower Gardener <gardener@tensorflow.org> | 2017-11-27 04:30:03 -0800 |
| commit | a264269f523467ac018708a647eab02c1f1010fe (patch) | |
| tree | d48a2395a2c84f4d88182f32b7ff1f47f685684b | |
| parent | 93bce00552ac70cc2c9b72e5742f9de87d72985a (diff) | |
Fixed a minor typo in FisherEstimator docstring.
PiperOrigin-RevId: 176999852
| -rw-r--r-- | tensorflow/contrib/kfac/python/ops/estimator.py | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/tensorflow/contrib/kfac/python/ops/estimator.py b/tensorflow/contrib/kfac/python/ops/estimator.py index c353f3592f..27ff951f16 100644 --- a/tensorflow/contrib/kfac/python/ops/estimator.py +++ b/tensorflow/contrib/kfac/python/ops/estimator.py @@ -95,7 +95,7 @@ class FisherEstimator(object): blocks, kronecker factors, and losses associated with the graph. estimation_mode: The type of estimator to use for the Fishers. Can be - 'gradients', 'empirical', 'curvature_propagation', or 'exact'. + 'gradients', 'empirical', 'curvature_prop', or 'exact'. (Default: 'gradients'). 'gradients' is the basic estimation approach from the original K-FAC paper. 'empirical' computes the 'empirical' Fisher information matrix (which uses the data's distribution for the |