"SparseSoftmaxCrossEntropyWithLogits"));
  // `gradient_checker` only works with model that returns only 1 tensor.
  // Although, `ops::SparseSoftmaxCrossEntropyWithLogits` returns 2 tensors, the
  // second tensor isn't needed for computing gradient so we could safely drop
  // it.
  outputs[0] = loss;
  backprop->Unref();
  return absl::OkStatus();
}

Status BiasAddModel(AbstractContext* ctx,

unless indicated otherwise.

1) Add the new version to the `VERSIONS` variable inside
   `tensorflow/tools/toolchains/python/python_repo.bzl`. \
   While this isn't necessary for running the updater, it is required for
   actually using the new version with Tensorflow.

2) In the `WORKSPACE` file, add the new version to the `python_versions`

  std::vector<AbstractTensorHandle*> temp_outputs(2);
  TF_RETURN_IF_ERROR(
      ops::IdentityN(ctx, inputs, absl::MakeSpan(temp_outputs), "IdentityN"));
  // Although, `ops::IdentityN` returns 2 tensors, the first tensor isn't needed
  // for computing gradient so we could safely drop it.
  outputs[0] = temp_outputs[1];
  temp_outputs[0]->Unref();
  return absl::OkStatus();
}

class CppGradients

    s->status = InvalidArgument("TF_WhileParams `cond_output` field isn't set");
    return false;
  }
  for (int i = 0; i < params.ninputs; ++i) {
    if (params.body_outputs[i].oper == nullptr) {
      s->status = InvalidArgument("TF_WhileParams `body_outputs[", i, "]` ",
                                  "field isn't set");
      return false;
    }
  }
  if (params.name == nullptr) {

    read = 0;
  } else if (!absl::SimpleAtoi(content_length->second, &read)) {
    TF_SetStatus(status, TF_UNKNOWN, "Could not get content-length header");
    return -1;
  }
  // `TF_OUT_OF_RANGE` isn't considered as an error. So we clear it here.
  TF_SetStatus(status, TF_OK, "");
  TF_VLog(1, "Successful read of %s @ %u of size: %u", path.c_str(), offset,
          read);
  stream.read(buffer, read);

      const std::function<void(BackwardFunction*)>& backward_function_deleter);

  // Returns true if `Accumulate` is active somewhere above on the stack and
  // there isn't an intervening PushState. This is useful for ordering
  // ForwardAccumulators, where more deeply nested accumulators should not see
  // computations from less deeply nested accumulators.

    which determines whether the SavedModel format stores the Keras model/layer
    call functions. The traced functions allow Keras to revive custom models and
    layers without the original class definition, but if this isn't required the
    tracing can be disabled with the added option.
*   The `tf.keras.mixed_precision` API is now non-experimental. The
    non-experimental API differs from the experimental API in several ways.

// `cond_output` and `body_outputs`.
//
// If `status` is OK, the caller must call either TF_FinishWhile or
// TF_AbortWhile on the returned TF_WhileParams. If `status` isn't OK, the
// returned TF_WhileParams is not valid, and the caller should not call
// TF_FinishWhile() or TF_AbortWhile().
//
// Missing functionality (TODO):
// - Gradients
// - Reference-type inputs