diff options
| author |  A. Unique TensorFlower <gardener@tensorflow.org> | 2017-02-01 11:23:07 -0800 |
|---|---|---|
| committer | TensorFlower Gardener <gardener@tensorflow.org> | 2017-02-01 11:43:10 -0800 |
| commit | 8fe32029f77b287a207c3bc3e0127fb35a3e23da (patch) | |
| tree | cb4e0e5434e21f4ba2a8e980610503b78b3337b9 /tensorflow/cc/ops | |
| parent | 287e845c52d0cb5fb7f5b64658ef513883f66fc5 (diff) | |
Add a mechanism for hiding, skipping, and modifying the generated op
functions for C++. A souped-up version of the hidden_ops mechanism in
Python, the intent is to use this for most or all of the client
languages, with a common list of changes to make in a common file and
per-language overrides.
Also:
* include the documentation for outputs in the generated comments
* several updates to C++ API to match Python
* fix C++ shape function for ConcatV2 now that we use it by default
* split op_gen_lib out of core:framework, since it is only used by
the op generators, and I don't want to add another proto to
mobile builds
Change: 146267344
Diffstat (limited to 'tensorflow/cc/ops')
| -rw-r--r-- | tensorflow/cc/ops/op_gen_overrides.pbtxt | 42 |
1 files changed, 42 insertions, 0 deletions
diff --git a/tensorflow/cc/ops/op_gen_overrides.pbtxt b/tensorflow/cc/ops/op_gen_overrides.pbtxt new file mode 100644 index 0000000000..79d3a04012 --- /dev/null +++ b/tensorflow/cc/ops/op_gen_overrides.pbtxt @@ -0,0 +1,42 @@ +# array_ops +op { name: "BroadcastArgs" rename_to: "BroadcastDynamicShape" } +op { name: "ConcatOffset" skip: true } # Maybe should just be hidden? +op { name: "Concat" skip: true } +op { name: "ConcatV2" rename_to: "Concat" } +op { name: "MirrorPadGrad" hide: true } +op { name: "Reverse" skip: true } +op { name: "ReverseV2" rename_to: "Reverse" } + +# candidate_sampling_ops +# control_flow_ops +# ctc_ops +# data_flow_ops +op { name: "FakeQueue" skip: true } + +# functional_ops +# image_ops +# io_ops +# linalg_ops +# logging_ops +# math_ops +op { name: "All" alias: "ReduceAll" input_rename: { from: "reduction_indices" to: "axis" } } +op { name: "Any" alias: "ReduceAny" input_rename: { from: "reduction_indices" to: "axis" } } +op { name: "Max" alias: "ReduceMax" input_rename: { from: "reduction_indices" to: "axis" } } +op { name: "Mean" alias: "ReduceMean" input_rename: { from: "reduction_indices" to: "axis" } } +op { name: "Min" alias: "ReduceMin" input_rename: { from: "reduction_indices" to: "axis" } } +op { name: "Prod" alias: "ReduceProd" input_rename: { from: "reduction_indices" to: "axis" } } +op { name: "Sum" alias: "ReduceSum" input_rename: { from: "reduction_indices" to: "axis" } } + +# nn_ops +op { name: "TopKV2" rename_to: "TopK" alias: "TopKV2" } # TODO(josh11b): delete "TopKV2" alias once users updated + +# parsing_ops +# random_ops +# script_ops +# sdca_ops +# state_ops +# sparse_ops +# string_ops +# user_ops +# training_ops +# word2vec deprecated ops |