def call(self, x):
      return x + 1

  model = AddOneModel()

  x_train = tf.constant([1, 2, 3, 4, 5], dtype=tf.float32)
  y_train = tf.constant([2, 3, 4, 5, 6], dtype=tf.float32)

  model.compile(optimizer='sgd', loss='mse')
  model.fit(x_train, y_train, epochs=1)

  path = tempdir + '/add_one_model'
  model.save(path)
  return path


class TensorflowToStableHLOTest(test.TestCase):

  );
}


def TF_XlaSparseDenseMatmulGradWithSgdAndStaticBufferSizeOp : TF_Op<"XlaSparseDenseMatmulGradWithSgdAndStaticBufferSize", [Pure]> {
  let summary = "A XLA op which performs the SGD optimizer update for the dense-sparse matrix multiplication.";

  let arguments = (ins
    TF_Int32Tensor:$row_pointers,
    TF_Int32Tensor:$sorted_sample_ids,
    TF_Int32Tensor:$sorted_token_ids,

  let results = (outs);
}

def TF_LoadTPUEmbeddingStochasticGradientDescentParametersOp : TF_Op<"LoadTPUEmbeddingStochasticGradientDescentParameters", [TF_MustExecute, TF_TPUEmbeddingReadEffect]> {
  let summary = "Load SGD embedding parameters.";

  let description = [{
An op that loads optimization parameters into HBM for embedding. Must be
preceded by a ConfigureTPUEmbeddingHost op that sets up the correct

    *   Added `tf.keras.optimizers.experimental.Optimizer`. The reworked
        optimizer gives more control over different phases of optimizer calls,
        and is easier to customize. We provide Adam, SGD, Adadelta, AdaGrad and
        RMSprop optimizers based on
        `tf.keras.optimizers.experimental.Optimizer`. Generally the new
        optimizers work in the same way as the old ones, but support new