return {
      'key': (tf.compat.v1.saved_model.signature_def_utils.build_signature_def(
          inputs={'x': tensor_info_x},
          outputs={'r': tensor_info_r},
          method_name='some_function'))
  }, tf.initializers.global_variables(), None


if __name__ == '__main__':
  common_v1.set_tf_options()

  return {
      'key': (tf.compat.v1.saved_model.signature_def_utils.build_signature_def(
          inputs={'x': tensor_info_x},
          outputs={'r': tensor_info_r},
          method_name='some_function'))
  }, tf.tables_initializer(), tf.get_collection(tf.GraphKeys.ASSET_FILEPATHS)


if __name__ == '__main__':
  common_v1.set_tf_options()

  return {
      'key': (tf.compat.v1.saved_model.signature_def_utils.build_signature_def(
          inputs={'x': tensor_info_x},
          outputs={'r': tensor_info_r},
          method_name='some_function'))
  }, tf.tables_initializer(), None


if __name__ == '__main__':
  common_v1.set_tf_options()

              'x': tensor_info_x,
              'y': tensor_info_y
          },
          outputs={
              's': tensor_info_s,
              't': tensor_info_t
          },
          method_name='some_function')),
      'key2': (tf.compat.v1.saved_model.signature_def_utils.build_signature_def(
          inputs={
              'a': tensor_info_y,
              'b': tensor_info_x,
          },
          outputs={

          'key': (
              tf.compat.v1.saved_model.signature_def_utils.build_signature_def(
                  inputs={'x': tensor_info_x},
                  outputs={'r': tensor_info_r},
                  method_name='some_function',
              )
          )
      },
      tf.initializers.global_variables(),
      None,
  )


if __name__ == '__main__':
  common_v1.set_tf_options()

Below is an example of a function operating on the TensorFlow dialect:

```mlir {.mlir}
/// This is a regular function, taking inputs by value and returning a new value.
/// The body is a regular CFG.
func some_function(%input : tensor<*xf32>) -> tensor<*xf32> {
  // TensorFlow operations are not variadic: this `tf.add` operation always
  // takes two inputs and returns a single output. This simplifies

=a.unusedTokens.length&&void 0===a.bigHour),null!=Object.isFrozen&&Object.isFrozen(e))return s;e._isValid=s}return e._isValid}function f(e){var a=c(NaN);return null!=e?L(Y(a),e):Y(a).userInvalidated=!0,a}n=Array.prototype.some?Array.prototype.some:function(e){for(var a=Object(this),t=a.length>>>0,s=0;s<t;s++)if(s in a&&e.call(this,a[s],s,a))return!0;return!1};var d=l.momentProperties=[];function k(e,a){var t,s,n;if(o(a._isAMomentObject)||(e._isAMomentObject=a._isAMomentObject),o(a._i)||(e._i=a._...