Add function gradient support to C API

Also, change the internal representation of TF_Function and
rename TF_GraphAddFunction to TF_GraphAddFunctionCopy to make it
clear that a copy of the function is added to the graph. Any
subsequent modifications to the function will not be reflected
in the copy added to the graph.

PiperOrigin-RevId: 169187793

1 files changed, 8 insertions, 0 deletions

diff --git a/tensorflow/c/c_test_util.h b/tensorflow/c/c_test_util.h
index 91f96b0e5d..9cfedf36e5 100644
--- a/tensorflow/c/c_test_util.h
+++ b/tensorflow/c/c_test_util.h
@@ -94,6 +94,14 @@ bool GetFunctionDef(TF_Function* func, tensorflow::FunctionDef* func_def);
 bool GetAttrValue(TF_Operation* oper, const char* attr_name,
                   tensorflow::AttrValue* attr_value, TF_Status* s);
 
+// Returns a sorted vector of std::pair<function_name, gradient_func> from
+// graph_def.library().gradient()
+std::vector<std::pair<string, string>> GetGradDefs(
+    const tensorflow::GraphDef& graph_def);
+
+// Returns a sorted vector of names contained in `grad_def`
+std::vector<string> GetFuncNames(const tensorflow::GraphDef& graph_def);
+
 class CSession {
  public:
   CSession(TF_Graph* graph, TF_Status* s);