const std::vector<float>& expected_values) {
  std::unique_ptr<TF_Status, decltype(&TF_DeleteStatus)> status(
      TF_NewStatus(), TF_DeleteStatus);
  TF_Tensor* t = TFE_TensorHandleResolve(handle, status.get());
  ASSERT_EQ(TF_OK, TF_GetCode(status.get())) << TF_Message(status.get());
  std::unique_ptr<float[]> actual_values(new float[expected_values.size()]);

    }
    if (op_type_.empty()) {
      return errors::FailedPrecondition(
          "GraphOperation::Reset must be called before calling SetOpName.");
    }
    // TODO(b/145674566): We use Graph::NewName to get a unique name here but
    // this may not be consistent with python's naming policy.
    mutex_lock l(g_->mu);
    op_.reset(new TF_OperationDescription(g_, op_type_.c_str(),

class OpDeleter {
 public:
  void operator()(TFE_Op* to_delete) const { TFE_DeleteOp(to_delete); }
};

using OpPtr = std::unique_ptr<TFE_Op, OpDeleter>;

class StatusDeleter {
 public:
  void operator()(TF_Status* to_delete) const { TF_DeleteStatus(to_delete); }
};

using StatusPtr = std::unique_ptr<TF_Status, StatusDeleter>;

class ExecutorDeleter {
 public:
  void operator()(TFE_Executor* to_delete) const {

      : owned_executor(nullptr), unowned_executor(executor) {}

  tensorflow::EagerExecutor* executor() {
    return owned_executor == nullptr ? unowned_executor : owned_executor.get();
  }

  std::unique_ptr<tensorflow::EagerExecutor> owned_executor;
  tensorflow::EagerExecutor* unowned_executor;
};

Variable* Variable::Create(TFE_Context* context, TF_DataType type,
                           const int64_t* dims, const int num_dims,
                           const char* device, TF_Status* status) {
  std::unique_ptr<TFE_Op, decltype(&TFE_DeleteOp)> op(
      TFE_NewOp(context, "VarHandleOp", status), TFE_DeleteOp);
  if (TF_GetCode(status) != TF_OK) return nullptr;
  TFE_OpSetAttrType(op.get(), "dtype", type);

  Tape tape(/*persistent=*/false);
  tape.Watch(inputs[0]);  // Watch x.
  AbstractTensorHandle* exp_output;
  TF_RETURN_IF_ERROR(ops::Exp(ctx, inputs[0], &exp_output, "Exp"));
  std::unique_ptr<GradientFunction> gradient_function(
      new PassThroughGradientFunction);
  tape.RecordOperation(inputs, {exp_output}, gradient_function.release());
  TF_RETURN_IF_ERROR(tape.ComputeGradient(ctx,

namespace parallel_device {
namespace {

class OpDeleter {
 public:
  void operator()(TFE_Op* to_delete) const { TFE_DeleteOp(to_delete); }
};

using OpPtr = std::unique_ptr<TFE_Op, OpDeleter>;

using MaybeParallelTensorOwned =
    absl::variant<std::unique_ptr<ParallelTensor>, TensorHandlePtr>;

using MaybeParallelTensorUnowned =
    absl::variant<ParallelTensor*, TFE_TensorHandle*>;

struct AbstractOperationDeleter {
  void operator()(AbstractOperation* p) const {
    if (p != nullptr) {
      p->Release();
    }
  }
};
}  // namespace internal

using AbstractOperationPtr =
    std::unique_ptr<AbstractOperation, internal::AbstractOperationDeleter>;

}  // namespace tensorflow

#include "tensorflow/core/framework/node_def.pb.h"
#include "tensorflow/core/framework/types.pb.h"
#include "tensorflow/core/platform/test.h"

using ::tensorflow::string;

typedef std::unique_ptr<TF_Tensor, decltype(&TF_DeleteTensor)>
    unique_tensor_ptr;

TF_Tensor* BoolTensor(int32_t v);

// Create a tensor with values of type TF_INT8 provided by `values`.
TF_Tensor* Int8Tensor(const int64_t* dims, int num_dims, const char* values);

  # Column 1 is left only, Column 2 is right only, Column 3 is shared lines
  # Select lines unique to actual_licenses, i.e. extra licenses.
  comm -1 -3 $BATS_TEST_TMPDIR/expected_licenses $BATS_TEST_TMPDIR/actual_licenses | grep -v -f $BATS_TEST_TMPDIR/allowed_to_be_extra > $BATS_TEST_TMPDIR/actual_extra_licenses || true
  # Select lines unique to expected_licenses, i.e. missing licenses