std::unique_ptr<CSession> expected_csession(
        new CSession(expected_graph_, s_));

    std::vector<TF_Output> grad_outputs_vec;
    grad_outputs_vec.assign(grad_outputs, grad_outputs + 2);
    csession->SetOutputs(grad_outputs_vec);
    csession->Run(s_);
    ASSERT_EQ(TF_OK, TF_GetCode(s_)) << TF_Message(s_);
    TF_Tensor* out0 = csession->output_tensor(0);

  reasonable contribution (meaning it is not a single liner cosmetic PR).

**2. Valid?**

-   If the PR passes all the quality checks then we go ahead and assign a
    reviewer.
-   If the PR didn't meet the validation criteria, we request for additional
    changes to be made to PR to pass quality checks and send it back or on a
    rare occasion we may reject it.

**3. Review**

    TFE_Context* ctx, TF_Tensor* t, TF_Status* status);

// Create a packed TensorHandle with the given list of TensorHandles.
// If `handles` are on the same device, assign the same device to the packed
// handle; if `handles` are on different deivces, assign a CompositeDevice to
// it.
TF_CAPI_EXPORT extern TFE_TensorHandle* TFE_CreatePackedTensorHandle(
    TFE_Context* ctx, TFE_TensorHandle** handles, int* num_handles,

      status->status = InvalidArgument("Unparseable RunOptions proto");
      return;
    }
    if (run_metadata != nullptr && run_metadata->data != nullptr) {
      status->status =
          InvalidArgument("Passing non-empty run_metadata is invalid.");
      return;
    }

    RunMetadata run_metadata_proto;
    result = session->Run(run_options_proto, input_pairs, output_tensor_names,

  TFE_Execute(op, &var_handle, &num_retvals, status);
  if (TF_GetCode(status) != TF_OK) return nullptr;
  TFE_DeleteOp(op);
  if (TF_GetCode(status) != TF_OK) return nullptr;
  CHECK_EQ(1, num_retvals);

  // Assign 'value' to it.
  op = TFE_NewOp(ctx, "AssignVariableOp", status);
  if (TF_GetCode(status) != TF_OK) return nullptr;
  TFE_OpSetAttrType(op, "dtype", TF_FLOAT);
  TFE_OpAddInput(op, var_handle, status);

        `tf.keras.mixed_precision.experimental.Policy`
    *   When passing a `tf.mixed_precision.experimental.DynamicLossScale` to
        `tf.keras.mixed_precision.experimental.LossScaleOptimizer`, the
        `DynamicLossScale`'s multiplier must be 2.
    *   When passing a `tf.mixed_precision.experimental.DynamicLossScale` to