def __init__()

in FasterRCNNDetection/data/iwildcam_dataset.py [0:0]

• 42 lines of code
• 15 McCabe index (conditional complexity)

    def __init__(self, root, ann_file, max_images=None):

        self.root = root
        print('Loading annotations from: ' + os.path.basename(ann_file))
        with open(ann_file) as data_file:
            ann_data = json.load(data_file)

        # set up the filenames and annotations
        self.impaths = np.array([aa['file_name'] for aa in ann_data['images']])
        self.image_ids = np.array([aa['id'] for aa in ann_data['images']])
        
        # This loop reads the bboxes and corresponding labels and assigns them
        # the correct image. Kind of slow at the moment...
        self.bboxes = [[] for _ in self.image_ids]
        self.labels = [[] for _ in self.image_ids]
        # To speed up the loop, creating mapping for image_id to list index 
        image_id_to_idx = {id:idx for idx, id in enumerate(self.image_ids)}
        for ann in ann_data['annotations']:
            idx = image_id_to_idx[ann['image_id']]
            #check that the image contains an animal, if not, don't append a box or label to the
            #image list
            if 'bbox' in ann:

                # Bboxes should have ('ymin', 'xmin', 'ymax', 'xmax') format
                self.bboxes[idx].append([ann['bbox'][1], ann['bbox'][0],
                                          ann['bbox'][1] + ann['bbox'][3],
                                          ann['bbox'][0] + ann['bbox'][2]])
                # Currently we take the label from the annotation file, non-consecutive-
                # label-support would be great
                self.labels[idx].append(ann['category_id'])
            else:
                #self.bboxes[idx].append([-1.,-1.,0.,0.])
                #self.labels[idx].append(30)
         
                self.bboxes[idx].append([])
                self.labels[idx].append(30)

        # load classes
        self.classes = np.unique(sorted([cat['id'] for cat in ann_data['categories']]))
        
        if opt.dataset == 'oneclass':
            self.class_names = ['Animal']
        else:
            self.class_names = ['' for _ in range(self.get_class_count())]
        
        for cat in ann_data['categories']:    
            self.class_names[cat['id']] = '{}'.format(cat['name'])
        
        # print out some stats
        print("The dataset has {} images containing {} classes".format(
                  len(self.image_ids),
                  len(self.classes)))
        
        if max_images is not None:
            print('Selecting a subset of {} images from training and validation'.
                format(max_images))
            self.impaths = self.impaths[:max_images]
            self.image_ids = self.image_ids[:max_images]
            self.bboxes = self.bboxes[:max_images]
            self.labels = self.labels[:max_images]
            
        if opt.dataset == 'oneclass':
            print('Merging all classes to one category')
            for ll in self.labels:
                for l_idx in range(len(ll)):
                    ll[l_idx] = 0