def face_detect_simulate()

in source/simulate/detector.py [0:0]

• 24 lines of code
• 5 McCabe index (conditional complexity)

    def face_detect_simulate(self):
        image_names = [f for f in os.listdir(self._faces_detect_root_dir) if f.startswith('test_')]
        image_names = sorted(image_names)

        for name in image_names:
            full_path = os.path.join(self._faces_detect_root_dir, name)
            print('Test image {}:'.format(full_path))

            # Step 1: read image and execute base64 encoding
            image_base64_enc = self.get_base64_encoding(full_path)

            # Step 2: send request to backend
            request_body = {
                "timestamp": str(time.time()),
                "request_id": 1242322,
                "image_base64_enc": image_base64_enc
            }

            t1 = time.time()
            response = requests.post(self._face_detect_url, data=json.dumps(request_body))
            t2 = time.time()
            print('Time cost = {}'.format(t2 - t1))

            # Step 3: visualization
            response = json.loads(response.text)
            print('Response = {}'.format(response))

            bbox_coords = np.array(response['bbox_coords'])
            bbox_scores = np.array(response['bbox_scores'])
            class_ids = np.zeros(shape=(bbox_scores.shape[0], ))
            print('bbox_coords.shape = {}'.format(bbox_coords.shape))
            print('bbox_scores.shape = {}'.format(bbox_scores.shape))

            self.visualize(full_path, bbox_coords, bbox_scores, class_ids, label_name='face')