diff options
author | Brennan Saeta <saeta@google.com> | 2018-07-05 10:38:42 -0700 |
---|---|---|
committer | TensorFlower Gardener <gardener@tensorflow.org> | 2018-07-05 10:41:55 -0700 |
commit | 14ef3b67a32279643d468274adb3a3019b0dd9c2 (patch) | |
tree | 27b5cf2e7a8a77afdbccb0156ce400283d71c596 /tensorflow/contrib/bigtable/BUILD | |
parent | baf41a35baddd6b637a7b9febf6acfa55c573168 (diff) |
[tf.data / Bigtable] Support sampling row keys.
When writing high performance input pipelines, you often need to read from multiple servers in parallel. Being able to sample row keys from a table allows one to easily construct high performance parallel input pipelines from Cloud Bigtable.
PiperOrigin-RevId: 203389017
Diffstat (limited to 'tensorflow/contrib/bigtable/BUILD')
-rw-r--r-- | tensorflow/contrib/bigtable/BUILD | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/tensorflow/contrib/bigtable/BUILD b/tensorflow/contrib/bigtable/BUILD index 037645aa6e..7c4f26d791 100644 --- a/tensorflow/contrib/bigtable/BUILD +++ b/tensorflow/contrib/bigtable/BUILD @@ -44,6 +44,7 @@ KERNEL_FILES = [ "kernels/bigtable_lookup_dataset_op.cc", "kernels/bigtable_prefix_key_dataset_op.cc", "kernels/bigtable_range_key_dataset_op.cc", + "kernels/bigtable_sample_keys_dataset_op.cc", "kernels/bigtable_scan_dataset_op.cc", ] |