diff options
| author |  A. Unique TensorFlower <gardener@tensorflow.org> | 2016-10-11 17:34:13 -0800 |
|---|---|---|
| committer | TensorFlower Gardener <gardener@tensorflow.org> | 2016-10-11 18:50:14 -0700 |
| commit | 2c183364e6994ef00d3cff930cb383c4e9443f25 (patch) | |
| tree | 95abd17465954d768d834f2afeea666f3f208f9f /tensorflow/contrib/training/__init__.py | |
| parent | 40fc65b0f5791f40132fbd173bfde7521f14fed5 (diff) | |
Check in code to perform rejection sampling.
Change: 135869416
Diffstat (limited to 'tensorflow/contrib/training/__init__.py')
| -rw-r--r-- | tensorflow/contrib/training/__init__.py | 11 |
1 files changed, 7 insertions, 4 deletions
diff --git a/tensorflow/contrib/training/__init__.py b/tensorflow/contrib/training/__init__.py index fc0e324bcf..1290854260 100644 --- a/tensorflow/contrib/training/__init__.py +++ b/tensorflow/contrib/training/__init__.py @@ -30,10 +30,12 @@ like to store state in the forward direction across segments of an example. ## Online data resampling To resample data with replacement on a per-example basis, use -['resample_at_rate'](#resample_at_rate), providing the desired rate -for each example. If you wish to specify relative rates, rather than -absolute ones, use ['weighted_resample'](#weighted_resample) (which -also returns the actual resampling rate used for each output example). +['rejection_sample'](#rejection_sample) or +['resample_at_rate'](#resample_at_rate). For `rejection_sample`, provide +a boolean Tensor describing whether to accept or reject. For `resample_at_rate`, +providing the desired rate for each example. If you wish to specify relative +rates, rather than absolute ones, use ['weighted_resample'](#weighted_resample) +(which also returns the actual resampling rate used for each output example). Use ['stratified_sample'](#stratified_sample) or ['stratified_sample_unknown_dist'](#stratified_sample_unknown_dist) to @@ -43,6 +45,7 @@ have a binary classification dataset that is 99.9% class 1, a common approach is to resample from the data so that the data is more balanced. +@@rejection_sample @@resample_at_rate @@stratified_sample @@stratified_sample_unknown_dist |