diff options
| author |  Taehoon Lee <me@taehoonlee.com> | 2018-07-09 17:25:20 +0900 |
|---|---|---|
| committer | Taehoon Lee <me@taehoonlee.com> | 2018-07-09 17:25:20 +0900 |
| commit | 61dbfba344abd770666842023b28e8a97a92adb5 (patch) | |
| tree | 0d63c3824894de0bb49d9e3c401062347f8e01f8 /tensorflow/contrib/mixed_precision | |
| parent | caf711b6be448d46354e0cbef23989fa837efcf7 (diff) | |
Fix typos
Diffstat (limited to 'tensorflow/contrib/mixed_precision')
| -rw-r--r-- | tensorflow/contrib/mixed_precision/python/loss_scale_optimizer.py | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/tensorflow/contrib/mixed_precision/python/loss_scale_optimizer.py b/tensorflow/contrib/mixed_precision/python/loss_scale_optimizer.py index ef34f7bf7b..93050a3ae3 100644 --- a/tensorflow/contrib/mixed_precision/python/loss_scale_optimizer.py +++ b/tensorflow/contrib/mixed_precision/python/loss_scale_optimizer.py @@ -77,7 +77,7 @@ class LossScaleOptimizer(optimizer.Optimizer): If gradients clipping is applied, one can call `optimizer.compute_gradients()` and `optimizer.apply_gradients()` - seperately. + separately. Notice the following way of using LossScaleOptimizer is not intended. Always use `loss_scale_optimizer.compute_gradients()` to compute gradients instead of |