Python object_detection.core.standard_fields.InputDataFields() Examples

def _get_features_dict(input_dict):
  """Extracts features dict from input dict."""

  source_id = _replace_empty_string_with_random_number(
      input_dict[fields.InputDataFields.source_id])

  hash_from_source_id = tf.string_to_hash_bucket_fast(source_id, HASH_BINS)
  features = {
      fields.InputDataFields.image:
          input_dict[fields.InputDataFields.image],
      HASH_KEY: tf.cast(hash_from_source_id, tf.int32),
      fields.InputDataFields.true_image_shape:
          input_dict[fields.InputDataFields.true_image_shape],
      fields.InputDataFields.original_image_spatial_shape:
          input_dict[fields.InputDataFields.original_image_spatial_shape]
  }
  if fields.InputDataFields.original_image in input_dict:
    features[fields.InputDataFields.original_image] = input_dict[
        fields.InputDataFields.original_image]
  return features 

def filter_groundtruth_with_nan_box_coordinates(tensor_dict):
  """Filters out groundtruth with no bounding boxes.

  Args:
    tensor_dict: a dictionary of following groundtruth tensors -
      fields.InputDataFields.groundtruth_boxes
      fields.InputDataFields.groundtruth_classes
      fields.InputDataFields.groundtruth_confidences
      fields.InputDataFields.groundtruth_keypoints
      fields.InputDataFields.groundtruth_instance_masks
      fields.InputDataFields.groundtruth_is_crowd
      fields.InputDataFields.groundtruth_area
      fields.InputDataFields.groundtruth_label_types

  Returns:
    a dictionary of tensors containing only the groundtruth that have bounding
    boxes.
  """
  groundtruth_boxes = tensor_dict[fields.InputDataFields.groundtruth_boxes]
  nan_indicator_vector = tf.greater(tf.reduce_sum(tf.cast(
      tf.is_nan(groundtruth_boxes), dtype=tf.int32), reduction_indices=[1]), 0)
  valid_indicator_vector = tf.logical_not(nan_indicator_vector)
  valid_indices = tf.where(valid_indicator_vector)

  return retain_groundtruth(tensor_dict, valid_indices) 

def test_value_error_on_duplicate_images(self):
    categories = [{'id': 1, 'name': 'cat'},
                  {'id': 2, 'name': 'dog'},
                  {'id': 3, 'name': 'elephant'}]
    #  Add groundtruth
    pascal_evaluator = object_detection_evaluation.PascalDetectionEvaluator(
        categories)
    image_key1 = 'img1'
    groundtruth_boxes1 = np.array([[0, 0, 1, 1], [0, 0, 2, 2], [0, 0, 3, 3]],
                                  dtype=float)
    groundtruth_class_labels1 = np.array([1, 3, 1], dtype=int)
    pascal_evaluator.add_single_ground_truth_image_info(
        image_key1,
        {standard_fields.InputDataFields.groundtruth_boxes: groundtruth_boxes1,
         standard_fields.InputDataFields.groundtruth_classes:
         groundtruth_class_labels1})
    with self.assertRaises(ValueError):
      pascal_evaluator.add_single_ground_truth_image_info(
          image_key1,
          {standard_fields.InputDataFields.groundtruth_boxes:
           groundtruth_boxes1,
           standard_fields.InputDataFields.groundtruth_classes:
           groundtruth_class_labels1}) 

def test_value_error_on_duplicate_images(self):
    #  Add groundtruth
    self.wp_eval = (
        object_detection_evaluation.PrecisionAtRecallDetectionEvaluator(
            self.categories, recall_lower_bound=0.0, recall_upper_bound=0.5))
    image_key1 = 'img1'
    groundtruth_boxes1 = np.array([[0, 0, 1, 1], [0, 0, 2, 2], [0, 0, 3, 3]],
                                  dtype=float)
    groundtruth_class_labels1 = np.array([1, 3, 1], dtype=int)
    self.wp_eval.add_single_ground_truth_image_info(
        image_key1, {
            standard_fields.InputDataFields.groundtruth_boxes:
                groundtruth_boxes1,
            standard_fields.InputDataFields.groundtruth_classes:
                groundtruth_class_labels1
        })
    with self.assertRaises(ValueError):
      self.wp_eval.add_single_ground_truth_image_info(
          image_key1, {
              standard_fields.InputDataFields.groundtruth_boxes:
                  groundtruth_boxes1,
              standard_fields.InputDataFields.groundtruth_classes:
                  groundtruth_class_labels1
          }) 

def create_and_add_common_ground_truth(self):
    #  Add groundtruth
    self.wp_eval = (
        object_detection_evaluation.WeightedPascalDetectionEvaluator(
            self.categories))

    image_key1 = 'img1'
    groundtruth_boxes1 = np.array([[0, 0, 1, 1], [0, 0, 2, 2], [0, 0, 3, 3]],
                                  dtype=float)
    groundtruth_class_labels1 = np.array([1, 3, 1], dtype=int)
    self.wp_eval.add_single_ground_truth_image_info(
        image_key1,
        {standard_fields.InputDataFields.groundtruth_boxes: groundtruth_boxes1,
         standard_fields.InputDataFields.groundtruth_classes:
         groundtruth_class_labels1})
    # add 'img2' separately
    image_key3 = 'img3'
    groundtruth_boxes3 = np.array([[0, 0, 1, 1]], dtype=float)
    groundtruth_class_labels3 = np.array([2], dtype=int)
    self.wp_eval.add_single_ground_truth_image_info(
        image_key3,
        {standard_fields.InputDataFields.groundtruth_boxes: groundtruth_boxes3,
         standard_fields.InputDataFields.groundtruth_classes:
         groundtruth_class_labels3}) 

def create_and_add_common_ground_truth(self):
    #  Add groundtruth
    self.wp_eval = (
        object_detection_evaluation.WeightedPascalDetectionEvaluator(
            self.categories))

    image_key1 = 'img1'
    groundtruth_boxes1 = np.array([[0, 0, 1, 1], [0, 0, 2, 2], [0, 0, 3, 3]],
                                  dtype=float)
    groundtruth_class_labels1 = np.array([1, 3, 1], dtype=int)
    self.wp_eval.add_single_ground_truth_image_info(
        image_key1,
        {standard_fields.InputDataFields.groundtruth_boxes: groundtruth_boxes1,
         standard_fields.InputDataFields.groundtruth_classes:
         groundtruth_class_labels1})
    # add 'img2' separately
    image_key3 = 'img3'
    groundtruth_boxes3 = np.array([[0, 0, 1, 1]], dtype=float)
    groundtruth_class_labels3 = np.array([2], dtype=int)
    self.wp_eval.add_single_ground_truth_image_info(
        image_key3,
        {standard_fields.InputDataFields.groundtruth_boxes: groundtruth_boxes3,
         standard_fields.InputDataFields.groundtruth_classes:
         groundtruth_class_labels3}) 

def test_value_error_on_duplicate_images(self):
    categories = [{'id': 1, 'name': 'cat'},
                  {'id': 2, 'name': 'dog'},
                  {'id': 3, 'name': 'elephant'}]
    #  Add groundtruth
    pascal_evaluator = object_detection_evaluation.PascalDetectionEvaluator(
        categories)
    image_key1 = 'img1'
    groundtruth_boxes1 = np.array([[0, 0, 1, 1], [0, 0, 2, 2], [0, 0, 3, 3]],
                                  dtype=float)
    groundtruth_class_labels1 = np.array([1, 3, 1], dtype=int)
    pascal_evaluator.add_single_ground_truth_image_info(
        image_key1,
        {standard_fields.InputDataFields.groundtruth_boxes: groundtruth_boxes1,
         standard_fields.InputDataFields.groundtruth_classes:
         groundtruth_class_labels1})
    with self.assertRaises(ValueError):
      pascal_evaluator.add_single_ground_truth_image_info(
          image_key1,
          {standard_fields.InputDataFields.groundtruth_boxes:
           groundtruth_boxes1,
           standard_fields.InputDataFields.groundtruth_classes:
           groundtruth_class_labels1}) 

def _get_features_dict(input_dict):
  """Extracts features dict from input dict."""

  source_id = _replace_empty_string_with_random_number(
      input_dict[fields.InputDataFields.source_id])

  hash_from_source_id = tf.string_to_hash_bucket_fast(source_id, HASH_BINS)
  features = {
      fields.InputDataFields.image:
          input_dict[fields.InputDataFields.image],
      HASH_KEY: tf.cast(hash_from_source_id, tf.int32),
      fields.InputDataFields.true_image_shape:
          input_dict[fields.InputDataFields.true_image_shape],
      fields.InputDataFields.original_image_spatial_shape:
          input_dict[fields.InputDataFields.original_image_spatial_shape]
  }
  if fields.InputDataFields.original_image in input_dict:
    features[fields.InputDataFields.original_image] = input_dict[
        fields.InputDataFields.original_image]
  return features 

def test_value_error_on_duplicate_images(self):
    #  Add groundtruth
    self.wp_eval = (
        object_detection_evaluation.WeightedPascalDetectionEvaluator(
            self.categories))
    image_key1 = 'img1'
    groundtruth_boxes1 = np.array([[0, 0, 1, 1], [0, 0, 2, 2], [0, 0, 3, 3]],
                                  dtype=float)
    groundtruth_class_labels1 = np.array([1, 3, 1], dtype=int)
    self.wp_eval.add_single_ground_truth_image_info(
        image_key1,
        {standard_fields.InputDataFields.groundtruth_boxes: groundtruth_boxes1,
         standard_fields.InputDataFields.groundtruth_classes:
         groundtruth_class_labels1})
    with self.assertRaises(ValueError):
      self.wp_eval.add_single_ground_truth_image_info(
          image_key1,
          {standard_fields.InputDataFields.groundtruth_boxes:
           groundtruth_boxes1,
           standard_fields.InputDataFields.groundtruth_classes:
           groundtruth_class_labels1}) 

def augment_input_data(tensor_dict, data_augmentation_options):
    """Applies data augmentation ops to input tensors.

    Args:
      tensor_dict: A dictionary of input tensors keyed by fields.InputDataFields.
      data_augmentation_options: A list of tuples, where each tuple contains a
        function and a dictionary that contains arguments and their values.
        Usually, this is the output of core/preprocessor.build.

    Returns:
      A dictionary of tensors obtained by applying data augmentation ops to the
      input tensor dictionary.
    """
    tensor_dict[fields.InputDataFields.image] = tf.expand_dims(
        tf.to_float(tensor_dict[fields.InputDataFields.image]), 0)

    include_instance_masks = (fields.InputDataFields.groundtruth_instance_masks
                              in tensor_dict)
    include_keypoints = (fields.InputDataFields.groundtruth_keypoints
                         in tensor_dict)
    tensor_dict = preprocessor.preprocess(
        tensor_dict, data_augmentation_options,
        func_arg_map=preprocessor.get_default_func_arg_map(
            include_instance_masks=include_instance_masks,
            include_keypoints=include_keypoints))
    tensor_dict[fields.InputDataFields.image] = tf.squeeze(
        tensor_dict[fields.InputDataFields.image], axis=0)
    return tensor_dict 

def test_returns_correct_metric_values_with_difficult_list(self):
    self.create_and_add_common_ground_truth()
    image_key2 = 'img2'
    groundtruth_boxes2 = np.array([[10, 10, 11, 11], [500, 500, 510, 510],
                                   [10, 10, 12, 12]], dtype=float)
    groundtruth_class_labels2 = np.array([1, 1, 3], dtype=int)
    groundtruth_is_difficult_list2 = np.array([False, True, False], dtype=bool)
    self.wp_eval.add_single_ground_truth_image_info(
        image_key2,
        {standard_fields.InputDataFields.groundtruth_boxes: groundtruth_boxes2,
         standard_fields.InputDataFields.groundtruth_classes:
         groundtruth_class_labels2,
         standard_fields.InputDataFields.groundtruth_difficult:
         groundtruth_is_difficult_list2
        })
    self.add_common_detected()

    metrics = self.wp_eval.evaluate()
    self.assertAlmostEqual(
        metrics[self.wp_eval._metric_prefix +
                'PerformanceByCategory/AP@0.5IOU/dog'], 0.0)
    self.assertAlmostEqual(
        metrics[self.wp_eval._metric_prefix +
                'PerformanceByCategory/AP@0.5IOU/elephant'], 0.0)
    self.assertAlmostEqual(
        metrics[self.wp_eval._metric_prefix +
                'PerformanceByCategory/AP@0.5IOU/cat'], 0.5 / 3)
    self.assertAlmostEqual(metrics[self.wp_eval._metric_prefix +
                                   'Precision/mAP@0.5IOU'],
                           1. / (3 + 1 + 2) / 3)
    self.wp_eval.clear()
    self.assertFalse(self.wp_eval._image_ids) 

def filter_unrecognized_classes(tensor_dict):
  """Filters out class labels that are not unrecognized by the labelmap.

  Decoder would parse unrecognized classes (not included in the labelmap) to
  a label of value -1. Such targets are unecessary for training, and causes
  issue for evaluation, due to labeling mapping logic. This function filters
  those labels out for both training and evaluation.

  Args:
    tensor_dict: dictionary containing input tensors keyed by
      fields.InputDataFields.

  Returns:
    A dictionary keyed by fields.InputDataFields containing the tensors
    obtained after applying the filtering.

  Raises:
    ValueError: If groundtruth_classes tensor is not in tensor_dict.
  """
  if fields.InputDataFields.groundtruth_classes not in tensor_dict:
    raise ValueError('`groundtruth classes` not in tensor_dict.')
  # Refer to tf_example_decoder for how unrecognized labels are handled.
  unrecognized_label = -1
  recognized_indices = tf.where(
      tf.greater(tensor_dict[fields.InputDataFields.groundtruth_classes],
                 unrecognized_label))

  return retain_groundtruth(tensor_dict, recognized_indices) 

def retain_groundtruth_with_positive_classes(tensor_dict):
  """Retains only groundtruth with positive class ids.

  Args:
    tensor_dict: a dictionary of following groundtruth tensors -
      fields.InputDataFields.groundtruth_boxes
      fields.InputDataFields.groundtruth_classes
      fields.InputDataFields.groundtruth_confidences
      fields.InputDataFields.groundtruth_keypoints
      fields.InputDataFields.groundtruth_instance_masks
      fields.InputDataFields.groundtruth_is_crowd
      fields.InputDataFields.groundtruth_area
      fields.InputDataFields.groundtruth_label_types
      fields.InputDataFields.groundtruth_difficult

  Returns:
    a dictionary of tensors containing only the groundtruth with positive
    classes.

  Raises:
    ValueError: If groundtruth_classes tensor is not in tensor_dict.
  """
  if fields.InputDataFields.groundtruth_classes not in tensor_dict:
    raise ValueError('`groundtruth classes` not in tensor_dict.')
  keep_indices = tf.where(tf.greater(
      tensor_dict[fields.InputDataFields.groundtruth_classes], 0))
  return retain_groundtruth(tensor_dict, keep_indices) 

def _get_groundtruth_data(detection_model, class_agnostic):
    """Extracts groundtruth data from detection_model.

    Args:
      detection_model: A `DetectionModel` object.
      class_agnostic: Whether the detections are class_agnostic.

    Returns:
      A tuple of:
      groundtruth: Dictionary with the following fields:
        'groundtruth_boxes': [num_boxes, 4] float32 tensor of boxes, in
          normalized coordinates.
        'groundtruth_classes': [num_boxes] int64 tensor of 1-indexed classes.
        'groundtruth_masks': 3D float32 tensor of instance masks (if provided in
          groundtruth)
      class_agnostic: Boolean indicating whether detections are class agnostic.
    """
    input_data_fields = fields.InputDataFields()
    groundtruth_boxes = detection_model.groundtruth_lists(
        fields.BoxListFields.boxes)[0]
    # For class-agnostic models, groundtruth one-hot encodings collapse to all
    # ones.
    if class_agnostic:
        groundtruth_boxes_shape = tf.shape(groundtruth_boxes)
        groundtruth_classes_one_hot = tf.ones([groundtruth_boxes_shape[0], 1])
    else:
        groundtruth_classes_one_hot = detection_model.groundtruth_lists(
            fields.BoxListFields.classes)[0]
    label_id_offset = 1  # Applying label id offset (b/63711816)
    groundtruth_classes = (
        tf.argmax(groundtruth_classes_one_hot, axis=1) + label_id_offset)
    groundtruth = {
        input_data_fields.groundtruth_boxes: groundtruth_boxes,
        input_data_fields.groundtruth_classes: groundtruth_classes
    }
    if detection_model.groundtruth_has_field(fields.BoxListFields.masks):
        groundtruth[input_data_fields.groundtruth_instance_masks] = (
            detection_model.groundtruth_lists(fields.BoxListFields.masks)[0])
    return groundtruth 

def test_returns_correct_metric_values_with_difficult_list(self):
    self.create_and_add_common_ground_truth()
    image_key2 = 'img2'
    groundtruth_boxes2 = np.array(
        [[10, 10, 11, 11], [500, 500, 510, 510], [10, 10, 12, 12]], dtype=float)
    groundtruth_class_labels2 = np.array([1, 1, 3], dtype=int)
    groundtruth_is_difficult_list2 = np.array([False, True, False], dtype=bool)
    self.wp_eval.add_single_ground_truth_image_info(
        image_key2, {
            standard_fields.InputDataFields.groundtruth_boxes:
                groundtruth_boxes2,
            standard_fields.InputDataFields.groundtruth_classes:
                groundtruth_class_labels2,
            standard_fields.InputDataFields.groundtruth_difficult:
                groundtruth_is_difficult_list2
        })
    self.add_common_detected()

    metrics = self.wp_eval.evaluate()
    self.assertAlmostEqual(
        metrics[self.wp_eval._metric_prefix +
                'PerformanceByCategory/AP@0.5IOU/dog'], 0.0)
    self.assertAlmostEqual(
        metrics[self.wp_eval._metric_prefix +
                'PerformanceByCategory/AP@0.5IOU/elephant'], 0.0)
    self.assertAlmostEqual(
        metrics[self.wp_eval._metric_prefix +
                'PerformanceByCategory/AP@0.5IOU/cat'], 0.5 / 3)
    self.assertAlmostEqual(
        metrics[self.wp_eval._metric_prefix +
                'Precision/mAP@0.5IOU@[0.0,0.5]Recall'], 1. / (3 + 1 + 2) / 3)
    self.wp_eval.clear()
    self.assertFalse(self.wp_eval._image_ids) 

def test_returns_correct_metric_values(self):
    self.create_and_add_common_ground_truth()
    image_key2 = 'img2'
    groundtruth_boxes2 = np.array([[10, 10, 11, 11], [500, 500, 510, 510],
                                   [10, 10, 12, 12]], dtype=float)
    groundtruth_class_labels2 = np.array([1, 1, 3], dtype=int)
    self.wp_eval.add_single_ground_truth_image_info(
        image_key2,
        {standard_fields.InputDataFields.groundtruth_boxes: groundtruth_boxes2,
         standard_fields.InputDataFields.groundtruth_classes:
         groundtruth_class_labels2
        })
    self.add_common_detected()

    metrics = self.wp_eval.evaluate()
    self.assertAlmostEqual(
        metrics[self.wp_eval._metric_prefix +
                'PerformanceByCategory/AP@0.5IOU/dog'], 0.0)
    self.assertAlmostEqual(
        metrics[self.wp_eval._metric_prefix +
                'PerformanceByCategory/AP@0.5IOU/elephant'], 0.0)
    self.assertAlmostEqual(
        metrics[self.wp_eval._metric_prefix +
                'PerformanceByCategory/AP@0.5IOU/cat'], 0.5 / 4)
    self.assertAlmostEqual(metrics[self.wp_eval._metric_prefix +
                                   'Precision/mAP@0.5IOU'],
                           1. / (4 + 1 + 2) / 3)
    self.wp_eval.clear()
    self.assertFalse(self.wp_eval._image_ids) 

def testRejectionOnDuplicateDetections(self):
    """Tests that detections cannot be added more than once for an image."""
    categories = [{'id': 1, 'name': 'cat'},
                  {'id': 2, 'name': 'dog'},
                  {'id': 3, 'name': 'elephant'}]
    #  Add groundtruth
    coco_evaluator = coco_evaluation.CocoDetectionEvaluator(categories)
    coco_evaluator.add_single_ground_truth_image_info(
        image_id='image1',
        groundtruth_dict={
            standard_fields.InputDataFields.groundtruth_boxes:
            np.array([[99., 100., 200., 200.]]),
            standard_fields.InputDataFields.groundtruth_classes: np.array([1])
        })
    coco_evaluator.add_single_detected_image_info(
        image_id='image1',
        detections_dict={
            standard_fields.DetectionResultFields.detection_boxes:
            np.array([[100., 100., 200., 200.]]),
            standard_fields.DetectionResultFields.detection_scores:
            np.array([.8]),
            standard_fields.DetectionResultFields.detection_classes:
            np.array([1])
        })
    detections_lists_len = len(coco_evaluator._detection_boxes_list)
    coco_evaluator.add_single_detected_image_info(
        image_id='image1',  # Note that this image id was previously added.
        detections_dict={
            standard_fields.DetectionResultFields.detection_boxes:
            np.array([[100., 100., 200., 200.]]),
            standard_fields.DetectionResultFields.detection_scores:
            np.array([.8]),
            standard_fields.DetectionResultFields.detection_classes:
            np.array([1])
        })
    self.assertEqual(detections_lists_len,
                     len(coco_evaluator._detection_boxes_list)) 

def augment_input_data(tensor_dict, data_augmentation_options):
  """Applies data augmentation ops to input tensors.

  Args:
    tensor_dict: A dictionary of input tensors keyed by fields.InputDataFields.
    data_augmentation_options: A list of tuples, where each tuple contains a
      function and a dictionary that contains arguments and their values.
      Usually, this is the output of core/preprocessor.build.

  Returns:
    A dictionary of tensors obtained by applying data augmentation ops to the
    input tensor dictionary.
  """
  tensor_dict[fields.InputDataFields.image] = tf.expand_dims(
      tf.to_float(tensor_dict[fields.InputDataFields.image]), 0)

  include_instance_masks = (fields.InputDataFields.groundtruth_instance_masks
                            in tensor_dict)
  include_keypoints = (fields.InputDataFields.groundtruth_keypoints
                       in tensor_dict)
  include_label_weights = (fields.InputDataFields.groundtruth_weights
                           in tensor_dict)
  include_label_confidences = (fields.InputDataFields.groundtruth_confidences
                               in tensor_dict)
  tensor_dict = preprocessor.preprocess(
      tensor_dict, data_augmentation_options,
      func_arg_map=preprocessor.get_default_func_arg_map(
          include_label_weights=include_label_weights,
          include_label_confidences=include_label_confidences,
          include_instance_masks=include_instance_masks,
          include_keypoints=include_keypoints))
  tensor_dict[fields.InputDataFields.image] = tf.squeeze(
      tensor_dict[fields.InputDataFields.image], axis=0)
  return tensor_dict 

def testRejectionOnDuplicateGroundtruth(self):
    """Tests that groundtruth cannot be added more than once for an image."""
    categories = [{'id': 1, 'name': 'cat'},
                  {'id': 2, 'name': 'dog'},
                  {'id': 3, 'name': 'elephant'}]
    #  Add groundtruth
    coco_evaluator = coco_evaluation.CocoDetectionEvaluator(categories)
    image_key1 = 'img1'
    groundtruth_boxes1 = np.array([[0, 0, 1, 1], [0, 0, 2, 2], [0, 0, 3, 3]],
                                  dtype=float)
    groundtruth_class_labels1 = np.array([1, 3, 1], dtype=int)
    coco_evaluator.add_single_ground_truth_image_info(image_key1, {
        standard_fields.InputDataFields.groundtruth_boxes:
            groundtruth_boxes1,
        standard_fields.InputDataFields.groundtruth_classes:
            groundtruth_class_labels1
    })
    groundtruth_lists_len = len(coco_evaluator._groundtruth_list)

    # Add groundtruth with the same image id.
    coco_evaluator.add_single_ground_truth_image_info(image_key1, {
        standard_fields.InputDataFields.groundtruth_boxes:
            groundtruth_boxes1,
        standard_fields.InputDataFields.groundtruth_classes:
            groundtruth_class_labels1
    })
    self.assertEqual(groundtruth_lists_len,
                     len(coco_evaluator._groundtruth_list)) 

def augment_input_data(tensor_dict, data_augmentation_options):
  """Applies data augmentation ops to input tensors.

  Args:
    tensor_dict: A dictionary of input tensors keyed by fields.InputDataFields.
    data_augmentation_options: A list of tuples, where each tuple contains a
      function and a dictionary that contains arguments and their values.
      Usually, this is the output of core/preprocessor.build.

  Returns:
    A dictionary of tensors obtained by applying data augmentation ops to the
    input tensor dictionary.
  """
  tensor_dict[fields.InputDataFields.image] = tf.expand_dims(
      tf.to_float(tensor_dict[fields.InputDataFields.image]), 0)

  include_instance_masks = (fields.InputDataFields.groundtruth_instance_masks
                            in tensor_dict)
  include_keypoints = (fields.InputDataFields.groundtruth_keypoints
                       in tensor_dict)
  tensor_dict = preprocessor.preprocess(
      tensor_dict, data_augmentation_options,
      func_arg_map=preprocessor.get_default_func_arg_map(
          include_instance_masks=include_instance_masks,
          include_keypoints=include_keypoints))
  tensor_dict[fields.InputDataFields.image] = tf.squeeze(
      tensor_dict[fields.InputDataFields.image], axis=0)
  return tensor_dict 

def test_returns_correct_metric_values(self):
    self.create_and_add_common_ground_truth()
    image_key2 = 'img2'
    groundtruth_boxes2 = np.array(
        [[10, 10, 11, 11], [500, 500, 510, 510], [10, 10, 12, 12]], dtype=float)
    groundtruth_class_labels2 = np.array([1, 1, 3], dtype=int)
    self.wp_eval.add_single_ground_truth_image_info(
        image_key2, {
            standard_fields.InputDataFields.groundtruth_boxes:
                groundtruth_boxes2,
            standard_fields.InputDataFields.groundtruth_classes:
                groundtruth_class_labels2
        })
    self.add_common_detected()

    metrics = self.wp_eval.evaluate()
    self.assertAlmostEqual(
        metrics[self.wp_eval._metric_prefix +
                'PerformanceByCategory/AP@0.5IOU/dog'], 0.0)
    self.assertAlmostEqual(
        metrics[self.wp_eval._metric_prefix +
                'PerformanceByCategory/AP@0.5IOU/elephant'], 0.0)
    self.assertAlmostEqual(
        metrics[self.wp_eval._metric_prefix +
                'PerformanceByCategory/AP@0.5IOU/cat'], 0.5 / 4)
    self.assertAlmostEqual(
        metrics[self.wp_eval._metric_prefix +
                'Precision/mAP@0.5IOU@[0.0,0.5]Recall'], 1. / (3 + 1 + 2) / 4)
    self.wp_eval.clear()
    self.assertFalse(self.wp_eval._image_ids) 

def testGetOneMAPWithMatchingGroundtruthAndDetectionsEmptyCrowd(self):
    """Tests computing mAP with empty is_crowd array passed in."""
    category_list = [{
        'id': 0,
        'name': 'person'
    }, {
        'id': 1,
        'name': 'cat'
    }, {
        'id': 2,
        'name': 'dog'
    }]
    coco_evaluator = coco_evaluation.CocoDetectionEvaluator(category_list)
    coco_evaluator.add_single_ground_truth_image_info(
        image_id='image1',
        groundtruth_dict={
            standard_fields.InputDataFields.groundtruth_boxes:
                np.array([[100., 100., 200., 200.]]),
            standard_fields.InputDataFields.groundtruth_classes:
                np.array([1]),
            standard_fields.InputDataFields.groundtruth_is_crowd:
                np.array([])
        })
    coco_evaluator.add_single_detected_image_info(
        image_id='image1',
        detections_dict={
            standard_fields.DetectionResultFields.detection_boxes:
                np.array([[100., 100., 200., 200.]]),
            standard_fields.DetectionResultFields.detection_scores:
                np.array([.8]),
            standard_fields.DetectionResultFields.detection_classes:
                np.array([1])
        })
    metrics = coco_evaluator.evaluate()
    self.assertAlmostEqual(metrics['DetectionBoxes_Precision/mAP'], 1.0) 

def _get_labels_dict(input_dict):
  """Extracts labels dict from input dict."""
  required_label_keys = [
      fields.InputDataFields.num_groundtruth_boxes,
      fields.InputDataFields.groundtruth_boxes,
      fields.InputDataFields.groundtruth_classes,
      fields.InputDataFields.groundtruth_weights,
  ]
  labels_dict = {}
  for key in required_label_keys:
    labels_dict[key] = input_dict[key]

  optional_label_keys = [
      fields.InputDataFields.groundtruth_confidences,
      fields.InputDataFields.groundtruth_keypoints,
      fields.InputDataFields.groundtruth_instance_masks,
      fields.InputDataFields.groundtruth_area,
      fields.InputDataFields.groundtruth_is_crowd,
      fields.InputDataFields.groundtruth_difficult
  ]

  for key in optional_label_keys:
    if key in input_dict:
      labels_dict[key] = input_dict[key]
  if fields.InputDataFields.groundtruth_difficult in labels_dict:
    labels_dict[fields.InputDataFields.groundtruth_difficult] = tf.cast(
        labels_dict[fields.InputDataFields.groundtruth_difficult], tf.int32)
  return labels_dict 

def _get_features_dict(input_dict):
  """Extracts features dict from input dict."""
  hash_from_source_id = tf.string_to_hash_bucket_fast(
      input_dict[fields.InputDataFields.source_id], HASH_BINS)
  features = {
      fields.InputDataFields.image:
          input_dict[fields.InputDataFields.image],
      HASH_KEY: tf.cast(hash_from_source_id, tf.int32),
      fields.InputDataFields.true_image_shape:
          input_dict[fields.InputDataFields.true_image_shape]
  }
  if fields.InputDataFields.original_image in input_dict:
    features[fields.InputDataFields.original_image] = input_dict[
        fields.InputDataFields.original_image]
  return features 

def _get_labels_dict(input_dict):
  """Extracts labels dict from input dict."""
  required_label_keys = [
      fields.InputDataFields.num_groundtruth_boxes,
      fields.InputDataFields.groundtruth_boxes,
      fields.InputDataFields.groundtruth_classes,
      fields.InputDataFields.groundtruth_weights
  ]
  labels_dict = {}
  for key in required_label_keys:
    labels_dict[key] = input_dict[key]

  optional_label_keys = [
      fields.InputDataFields.groundtruth_keypoints,
      fields.InputDataFields.groundtruth_instance_masks,
      fields.InputDataFields.groundtruth_area,
      fields.InputDataFields.groundtruth_is_crowd,
      fields.InputDataFields.groundtruth_difficult
  ]

  for key in optional_label_keys:
    if key in input_dict:
      labels_dict[key] = input_dict[key]
  if fields.InputDataFields.groundtruth_difficult in labels_dict:
    labels_dict[fields.InputDataFields.groundtruth_difficult] = tf.cast(
        labels_dict[fields.InputDataFields.groundtruth_difficult], tf.int32)
  return labels_dict 

def augment_input_data(tensor_dict, data_augmentation_options):
  """Applies data augmentation ops to input tensors.

  Args:
    tensor_dict: A dictionary of input tensors keyed by fields.InputDataFields.
    data_augmentation_options: A list of tuples, where each tuple contains a
      function and a dictionary that contains arguments and their values.
      Usually, this is the output of core/preprocessor.build.

  Returns:
    A dictionary of tensors obtained by applying data augmentation ops to the
    input tensor dictionary.
  """
  tensor_dict[fields.InputDataFields.image] = tf.expand_dims(
      tf.to_float(tensor_dict[fields.InputDataFields.image]), 0)

  include_instance_masks = (fields.InputDataFields.groundtruth_instance_masks
                            in tensor_dict)
  include_keypoints = (fields.InputDataFields.groundtruth_keypoints
                       in tensor_dict)
  tensor_dict = preprocessor.preprocess(
      tensor_dict, data_augmentation_options,
      func_arg_map=preprocessor.get_default_func_arg_map(
          include_instance_masks=include_instance_masks,
          include_keypoints=include_keypoints))
  tensor_dict[fields.InputDataFields.image] = tf.squeeze(
      tensor_dict[fields.InputDataFields.image], axis=0)
  return tensor_dict 

def _make_evaluation_dict(self):
    input_data_fields = fields.InputDataFields
    detection_fields = fields.DetectionResultFields

    image = tf.zeros(shape=[1, 20, 20, 3], dtype=tf.uint8)
    key = tf.constant('image1')
    detection_boxes = tf.constant([[[0., 0., 1., 1.]]])
    detection_scores = tf.constant([[0.8]])
    detection_classes = tf.constant([[0]])
    detection_masks = tf.ones(shape=[1, 1, 20, 20], dtype=tf.float32)
    num_detections = tf.constant([1])
    groundtruth_boxes = tf.constant([[0., 0., 1., 1.]])
    groundtruth_classes = tf.constant([1])
    groundtruth_instance_masks = tf.ones(shape=[1, 20, 20], dtype=tf.uint8)
    detections = {
        detection_fields.detection_boxes: detection_boxes,
        detection_fields.detection_scores: detection_scores,
        detection_fields.detection_classes: detection_classes,
        detection_fields.detection_masks: detection_masks,
        detection_fields.num_detections: num_detections
    }
    groundtruth = {
        input_data_fields.groundtruth_boxes: groundtruth_boxes,
        input_data_fields.groundtruth_classes: groundtruth_classes,
        input_data_fields.groundtruth_instance_masks: groundtruth_instance_masks
    }
    return eval_util.result_dict_for_single_example(image, key, detections,
                                                    groundtruth) 

def _get_features_dict(input_dict):
  """Extracts features dict from input dict."""
  hash_from_source_id = tf.string_to_hash_bucket_fast(
      input_dict[fields.InputDataFields.source_id], HASH_BINS)
  features = {
      fields.InputDataFields.image:
          input_dict[fields.InputDataFields.image],
      HASH_KEY: tf.cast(hash_from_source_id, tf.int32),
      fields.InputDataFields.true_image_shape:
          input_dict[fields.InputDataFields.true_image_shape]
  }
  if fields.InputDataFields.original_image in input_dict:
    features[fields.InputDataFields.original_image] = input_dict[
        fields.InputDataFields.original_image]
  return features 

def _get_labels_dict(input_dict):
  """Extracts labels dict from input dict."""
  required_label_keys = [
      fields.InputDataFields.num_groundtruth_boxes,
      fields.InputDataFields.groundtruth_boxes,
      fields.InputDataFields.groundtruth_classes,
      fields.InputDataFields.groundtruth_weights
  ]
  labels_dict = {}
  for key in required_label_keys:
    labels_dict[key] = input_dict[key]

  optional_label_keys = [
      fields.InputDataFields.groundtruth_keypoints,
      fields.InputDataFields.groundtruth_instance_masks,
      fields.InputDataFields.groundtruth_area,
      fields.InputDataFields.groundtruth_is_crowd,
      fields.InputDataFields.groundtruth_difficult
  ]

  for key in optional_label_keys:
    if key in input_dict:
      labels_dict[key] = input_dict[key]
  if fields.InputDataFields.groundtruth_difficult in labels_dict:
    labels_dict[fields.InputDataFields.groundtruth_difficult] = tf.cast(
        labels_dict[fields.InputDataFields.groundtruth_difficult], tf.int32)
  return labels_dict 

def _make_evaluation_dict(self):
    input_data_fields = fields.InputDataFields
    detection_fields = fields.DetectionResultFields

    image = tf.zeros(shape=[1, 20, 20, 3], dtype=tf.uint8)
    key = tf.constant('image1')
    detection_boxes = tf.constant([[[0., 0., 1., 1.]]])
    detection_scores = tf.constant([[0.8]])
    detection_classes = tf.constant([[0]])
    detection_masks = tf.ones(shape=[1, 1, 20, 20], dtype=tf.float32)
    num_detections = tf.constant([1])
    groundtruth_boxes = tf.constant([[0., 0., 1., 1.]])
    groundtruth_classes = tf.constant([1])
    groundtruth_instance_masks = tf.ones(shape=[1, 20, 20], dtype=tf.uint8)
    detections = {
        detection_fields.detection_boxes: detection_boxes,
        detection_fields.detection_scores: detection_scores,
        detection_fields.detection_classes: detection_classes,
        detection_fields.detection_masks: detection_masks,
        detection_fields.num_detections: num_detections
    }
    groundtruth = {
        input_data_fields.groundtruth_boxes: groundtruth_boxes,
        input_data_fields.groundtruth_classes: groundtruth_classes,
        input_data_fields.groundtruth_instance_masks: groundtruth_instance_masks
    }
    return eval_util.result_dict_for_single_example(image, key, detections,
                                                    groundtruth)