diff options
| author |  ngc92 <ngc92@users.noreply.github.com> | 2017-12-26 20:23:05 +0100 |
|---|---|---|
| committer |  drpngx <drpngx@users.noreply.github.com> | 2017-12-26 11:23:05 -0800 |
| commit | b1ffd57fde26dce0aca1f8f2243a9a4a2abb994a (patch) | |
| tree | 0c0823dfd8e7cef44f77c98ee98df8d264e6bc92 | |
| parent | d72d5b07bfd212adbfadb748c3dab5d56ec2fb9f (diff) | |
added note about weights gradient in compute_weighted_loss (#15533)
| -rw-r--r-- | tensorflow/python/ops/losses/losses_impl.py | 7 |
1 files changed, 7 insertions, 0 deletions
diff --git a/tensorflow/python/ops/losses/losses_impl.py b/tensorflow/python/ops/losses/losses_impl.py index b74971f654..4f24c3c5bf 100644 --- a/tensorflow/python/ops/losses/losses_impl.py +++ b/tensorflow/python/ops/losses/losses_impl.py @@ -157,6 +157,13 @@ def compute_weighted_loss( ValueError: If `weights` is `None` or the shape is not compatible with `losses`, or if the number of dimensions (rank) of either `losses` or `weights` is missing. + + Note: + When calculating the gradient of a weighted loss contributions from + both `losses` and `weights` are considered. If your `weights` depend + on some model parameters but you do not want this to affect the loss + gradient, you need to apply @{tf.stop_gradient} to `weights` before + passing them to `compute_weighted_loss`. """ Reduction.validate(reduction) with ops.name_scope(scope, "weighted_loss", (losses, weights)): |