output->DebugString(100));
}

REGISTER_OP("AllocateOutputOp0").Output("output1: float");

TEST_F(DeviceKernelOpTest, TestAllocateEmptyOutput) {
  auto my_compute_func = [](void* kernel, TF_OpKernelContext* ctx) {
    TF_Status* s = TF_NewStatus();
    // Allocate empty output
    int64_t dim = 0;
    TF_Tensor* output = TF_AllocateOutput(

overlapping-outputs-timeline.png...

overlapping-outputs-input-comparison.png...

      )

      inputs = {'input_vocabs': input_vocabs_placeholder}
      outputs = {
          'lookup': lookup_tensor,
          'output': output_tensor,
      }

      self._save_tf1_model(
          sess,
          output_path,
          signature_def_key,
          tags,
          inputs=inputs,
          outputs=outputs,
          init_op=lookup_ops.tables_initializer(),

//  noutputs - number of elements in `outputs` array
//  outputs - array of TF_Outputs that specify the outputs of the function.
//            If `noutputs` is zero (the function returns no outputs), `outputs`
//            can be null. `outputs` can contain the same tensor more than once.
//  output_names - The names of the function's outputs. `output_names` array

  for (const auto& op : subgraph.operators) {
    for (int32_t output : op->outputs) {
      if (!llvm::is_contained(input_indices, output)) {
        defining_op[output] = op.get();
      }
    }
  }

  std::vector<const tflite::OperatorT*> queue;
  for (int32_t output : output_indices) {
    if (auto& op = defining_op[output]) {
      queue.push_back(op);
    }
  }

        output.contains("Appending output property name to build cache key: outputDirectory")
        output.contains("Appending output property name to build cache key: resultsFile")
        output.contains("Build cache key for MakeGreen")
    }

    def "honors @PathSensitive(NONE) on input artifact property for project artifact file when caching"() {
        createDirs("a", "b", "c")
        settingsFile << "include 'a', 'b', 'c'"

  // Check that we have two inputs and one output.
  ASSERT_THAT(op->inputs, SizeIs(2));
  ASSERT_THAT(op->outputs, SizeIs(1));

  // Check that all is quantized.
  auto output = subgraph->tensors[op->outputs[0]].get();
  auto input1 = subgraph->tensors[op->inputs[0]].get();
  auto input2 = subgraph->tensors[op->inputs[1]].get();

  EXPECT_THAT(output->type, Eq(TensorType_INT8));

      [DeclareOpInterfaceMethods<SymbolUserOpInterface>]> {
  let summary = [{
output = input; While (Cond(output)) { output = Body(output) }
  }];

  let description = [{
output = input; While (Cond(output)) { output = Body(output) }

input: A list of input tensors whose types are T.
output: A list of output tensors whose types are T.
cond: A function that takes 'input' and returns a tensor.  If the tensor is

                         const std::vector<IOSpec>& outputs) {
    const OpDef& signature = fdef.signature();
    ASSERT_EQ(outputs.size(), signature.output_arg_size());
    for (int i = 0; i < outputs.size(); ++i) {
      const OpDef::ArgDef& arg = signature.output_arg(i);
      const IOSpec& out = outputs[i];
      if (out.second != DT_INVALID) {
        ASSERT_EQ(arg.type(), out.second)