return {'output': out}

    model = MatmulModel(weight_shape)
    saved_model_save.save(
        model,
        saved_model_path,
        signatures=model.matmul.get_concrete_function(
            tensor_spec.TensorSpec(
                shape=input_shape, dtype=dtypes.float32, name='input_tensor'
            )
        ),
    )
    return model

  def _any_log_contains(

    # Create and save a model having 2 signatures.
    model = MultipleSignatureModel()

    signatures = {
        'sig1': model.matmul.get_concrete_function(
            tensor_spec.TensorSpec(shape=(1, 4), dtype=dtypes.float32)
        ),
        'sig2': model.conv.get_concrete_function(
            tensor_spec.TensorSpec(shape=(1, 3, 4, 3), dtype=dtypes.float32)
        ),
    }
    saved_model_save.save(

        return {'output': out}

    model = TwoMatmulModel()
    input_shape = (1, 2)

    save.save(
        model,
        self._input_saved_model_path,
        signatures=model.matmul.get_concrete_function(
            tensor_spec.TensorSpec(
                shape=input_shape, dtype=dtypes.float32, name='input_tensor'
            )
        ),
    )

    def data_gen() -> repr_dataset.RepresentativeDataset:

        sometimes incorrectly ignored.
    *   functions returned by the `get_concrete_function` method of
        `tf.function` objects can now be called with arguments consistent with
        the original arguments or type specs passed to `get_concrete_function`.
        This calling convention is now the preferred way to use concrete

   return %ret : tensor<10x10xf32>
}
'''

@tf.function
def foo(x, y):
  return mlir_passthrough_op([x, y], mlir_module, Toutputs=[tf.float32])

graph_def = foo.get_concrete_function(tf.TensorSpec([10], tf.float32), tf.TensorSpec([10], tf.float32)).graph.as_graph_def()
```
  }];

  let arguments = (ins
    Variadic<TF_Tensor>:$inputs,

    StrAttr:$mlir_module