Function make_identity_tests

tensorflow/lite/testing/generate_examples_lib.py:829–864

Make a set of tests to do identity.

(options)

@register_make_test_function()
def make_identity_tests(options):
  """Make a set of tests to do identity."""

  # Chose a set of parameters
  test_parameters = [{
      "input_shape": [[], [1], [3, 3]],
      "op_to_use": ["identity", "identity_n", "snapshot"],
  }]

  def build_graph(parameters):
    input_tensor = tf.placeholder(
        dtype=tf.float32, name="input", shape=parameters["input_shape"])
    # We add the Multiply before Identity just as a walk-around to make the test
    # pass when input_shape is scalar.
    # During graph transformation, TOCO will replace the Identity op with
    # Reshape when input has shape. However, currently TOCO can't distinguish
    # between missing shape and scalar shape. As a result, when input has scalar
    # shape, this conversion still fails.
    # TODO(b/129197312), remove the walk-around code once the bug is fixed.
    input_doubled = input_tensor * 2.0
    if parameters["op_to_use"] == "identity":
      identity_output = tf.identity(input_doubled)
    elif parameters["op_to_use"] == "identity_n":
      # Testing `IdentityN` with a single tensor.
      identity_output = tf.identity_n([input_doubled])[0]
    elif parameters["op_to_use"] == "snapshot":
      identity_output = array_ops.snapshot(input_doubled)
    return [input_tensor], [identity_output]

  def build_inputs(parameters, sess, inputs, outputs):
    input_values = create_tensor_data(
        np.float32, parameters["input_shape"], min_value=-4, max_value=10)
    return [input_values], sess.run(
        outputs, feed_dict=dict(zip(inputs, [input_values])))

  make_zip_of_tests(options, test_parameters, build_graph, build_inputs)


@register_make_test_function()