Upload get_coordinate.py with huggingface_hub

get_coordinate.py

@@ -0,0 +1,112 @@
+# %%
+import json
+import os
+import sys
+import cv2
+import logging
+from logging.handlers import RotatingFileHandler
+from concurrent.futures import ThreadPoolExecutor
+from tqdm import tqdm
+
+# 配置日志记录到文件
+# 日志配置
+def setup_logger(log_file):
+    logger = logging.getLogger("ImageProcessingLogger")
+    logger.setLevel(logging.DEBUG)
+
+    # 设置文件滚动，每个日志文件最大10MB，保留5个文件
+    handler = RotatingFileHandler(log_file, maxBytes=10 * 1024 * 1024, backupCount=5)
+    handler.setLevel(logging.DEBUG)
+
+    # 设置日志格式
+    formatter = logging.Formatter('%(asctime)s - %(levelname)s - %(message)s')
+    handler.setFormatter(formatter)
+
+    # 添加处理器
+    logger.addHandler(handler)
+    return logger
+
+# 日志实例
+logger = setup_logger("app.log")
+
+
+def read_json(file_path): 
+    with open(file_path, 'r', encoding='utf-8') as file:
+        data = json.load(file)
+    return data
+
+def write_json(file_path, data):
+    with open(file_path, 'w', encoding='utf-8') as file:
+        json.dump(data, file, ensure_ascii=False, indent=4)
+
+
+
+def get_corrdinate(data, image_path,logger, out_root=None):
+    try:
+        image_name = image_path.split('/')[-1]
+        out_path = os.path.join(out_root, image_name)
+
+        y_str, x_str = data['click_loc'].split(',')[2:]
+        y = float(y_str.split(':')[1])
+        x = float(x_str.strip())
+        # print(f"x: {x}, y: {y}")
+
+        image = cv2.imread(image_path)
+        if image is None:
+            logger.error(f"Failed to read image: {image_path}")
+            return
+        height, width, _ = image.shape
+        # print(f"height: {height}, width: {width}")
+
+        x = int(x * width) 
+        y = int(y * height) 
+        # print(f"x: {x}, y: {y}")
+
+        color = (0, 255, 0)
+
+        thickness = 5
+        center = (x, y)
+        length = 35
+        cv2.line(image, (center[0] - length, center[1] - length), (center[0] - length, center[1] + length), color, thickness)
+        cv2.line(image, (center[0] - length, center[1] + length), (center[0] + length, center[1] + length), color, thickness)
+        cv2.line(image, (center[0] + length, center[1] + length), (center[0] + length, center[1] - length), color, thickness)
+        cv2.line(image, (center[0] + length, center[1] - length), (center[0] - length, center[1] - length), color, thickness)
+
+        cv2.imwrite(out_path, image)
+        
+        # 每处理100张记录日志
+        if count % 100 == 0:
+            logger.info(f"Processed {count} images successfully.")
+    except Exception as e:
+        logger.exception(f"Error processing data: {data}. Exception: {e}")
+
+
+def process_data(data, root_path,logger, out_root):
+    image_path = os.path.join(root_path, data['image'])
+    get_corrdinate(data, image_path,logger, out_root)
+
+
+if __name__ == "__main__":
+    # 参数配置
+    json_file = r'/code/Auto-GUI/dataset/mind/general_blip_train_llava_coco.json'
+    root_path = r'/code/Auto-GUI/dataset'
+    out_root = r'/code/Auto-GUI/dataset/coco_corrdinate'
+    count = 0
+
+
+    # 读取 JSON 数据
+    data = read_json(json_file)
+    data1 = [line for line in data if line['action_type'] == '#DUAL_POINT#'][1:]
+    logging.debug(f"total data: " +str(len(data1)))
+
+    # 并行处理数据
+    os.makedirs(out_root, exist_ok=True)
+    with ThreadPoolExecutor(max_workers=8) as executor:
+        list(tqdm(executor.map(lambda d: process_data(d, root_path, logger, out_root), data1), total=len(data1)))
+
+        # list(tqdm(executor.map(lambda d: process_data(d, root_path, out_root), data1), total=len(data1)))
+
+    logger.info("All processing complete.")
+
+
+