programs that TensorFlow executes. TensorFlow programs are encoded as
computation
[**graphs**](https://developers.google.com/machine-learning/glossary/#graph).
Since models are practically programs that TensorFlow executes, using untrusted
models or graphs is equivalent to running untrusted code.

If you need to run untrusted models, execute them inside a

    ],
    "access_right": "open",
    "notes": "Specific TensorFlow versions can be found in the \"Versions\" list on the right side of this page.<br>See the full list of authors <a href=\"https://github.com/tensorflow/tensorflow/graphs/contributors\">on GitHub</a>."

![Grafana](https://raw.githubusercontent.com/minio/minio/master/docs/metrics/prometheus/grafana/node/grafana-node.png)

  mutable tensorflow::mutex mu;
  tensorflow::Graph graph TF_GUARDED_BY(mu);

  // Runs shape inference.
  tensorflow::ShapeRefiner refiner TF_GUARDED_BY(mu);

  // Maps from name of an operation to the Node* in 'graph'.
  std::unordered_map<tensorflow::string, tensorflow::Node*> name_map
      TF_GUARDED_BY(mu);

  // The keys of this map are all the active sessions using this graph. Each

// The new graph construction API, still under development.

// Represents a computation graph.  Graphs may be shared between sessions.
// Graphs are thread-safe when used as directed below.
typedef struct TF_Graph TF_Graph;

// Return a new graph object.
TF_CAPI_EXPORT extern TF_Graph* TF_NewGraph(void);

// Destroy an options object. Graph will be deleted once no more

abstract: TensorFlow is a machine learning system that operates at large scale and in heterogeneous environments. TensorFlow uses dataflow graphs to represent computation, shared state, and the operations that mutate that state. It maps the nodes of a dataflow graph across many machines in a cluster, and within a machine across multiple computational devices, including multicore CPUs, general purpose GPUs, and custom-designed ASICs known as Tensor Processing...

            PerformanceTest(
                model,
                stage,
                flameGraphGenerationBuildSpec,
                description =
                    "Flame graphs with $profiler for ${performanceScenario.scenario.scenario} | ${performanceScenario.testProject} " +
                        "on ${os.asName()} (bucket $bucketIndex)",
                performanceSubProject = "performance",

  TF_Status* s = TF_NewStatus();
  TF_Graph* graph = TF_NewGraph();

  // Create a simple graph.
  Placeholder(graph, s);
  ASSERT_EQ(TF_OK, TF_GetCode(s)) << TF_Message(s);
  ASSERT_TRUE(TF_GraphOperationByName(graph, "feed") != nullptr);
  TF_Operation* oper = ScalarConst(3, graph, s);
  ASSERT_EQ(TF_OK, TF_GetCode(s)) << TF_Message(s);
  ASSERT_TRUE(TF_GraphOperationByName(graph, "scalar") != nullptr);
  Neg(oper, graph, s);

    have resulted in an error. When this happens, a noop is returned and the
    input tensors are marked non-feedable. In other words, if they are used as
    keys in `feed_dict` argument to `session.run()`, an error will be raised.
    Also, because some assert ops don't make it into the graph, the graph
    structure changes. A different graph can result in different per-op random

  TF_Graph* const graph = s->graph;
  if (graph != nullptr) {
    graph->mu.lock();
    graph->sessions.erase(s);
    const bool del = graph->delete_requested && graph->sessions.empty();
    graph->mu.unlock();
    if (del) delete graph;
  }
  delete s->session;
  delete s;
}

void TF_SessionRun(TF_Session* session, const TF_Buffer* run_options,