import os import json import re # 提取文件名中起始的连续数字(核心逻辑) def extract_starting_numbers(filename): # 匹配字符串开头的连续数字 match = re.match(r'^\d+', filename) if match: return match.group() # 返回起始数字字符串(如"4592") return filename # 无起始数字则返回原文件名(小写处理后) # 处理单个ldr文件数据 def process_ldr_data(lines, label_mapping, label_inverse_mapping, label_counter): all_labels = [] for line in lines: if line.startswith('1'): # 只处理零件数据行 parts = line.split() if len(parts) < 15: print(f"Skipping invalid line: {line.strip()}") continue # 1. 统一转为小写 filename = parts[14].lower() # 2. 提取起始连续数字作为标识 part_identifier = extract_starting_numbers(filename) print(f"Processing file: {filename} → Identifier: {part_identifier}") if part_identifier not in label_mapping: label_mapping[part_identifier] = label_counter label_inverse_mapping[label_counter] = part_identifier label_counter += 1 all_labels.append(label_mapping[part_identifier]) return label_mapping, label_inverse_mapping, label_counter # 处理一个文件夹中的所有ldr文件 def process_all_ldr_in_folder(folder_path): overall_label_mapping = {} overall_label_inverse_mapping = {} label_counter = 0 for root, dirs, files in os.walk(folder_path): for file in files: if file.endswith('.ldr'): file_path = os.path.join(root, file) with open(file_path, 'r') as f: lines = f.readlines() overall_label_mapping, overall_label_inverse_mapping, label_counter = process_ldr_data( lines, overall_label_mapping, overall_label_inverse_mapping, label_counter) return overall_label_mapping, overall_label_inverse_mapping # 保存mapping到文件 def save_mappings(label_mapping, label_inverse_mapping, output_dir): os.makedirs(output_dir, exist_ok=True) with open(os.path.join(output_dir, 'label_mapping.json'), 'w') as f: json.dump(label_mapping, f, indent=4) with open(os.path.join(output_dir, 'label_inverse_mapping.json'), 'w') as f: json.dump(label_inverse_mapping, f, indent=4) if __name__ == "__main__": input_folder = '/public/home/wangshuo/gap/assembly/data/car_1k/subset_bottom_300/ldr_rot_expand_trans' output_folder = '/public/home/wangshuo/gap/assembly/data/car_1k/subset_bottom_300' label_mapping, label_inverse_mapping = process_all_ldr_in_folder(input_folder) save_mappings(label_mapping, label_inverse_mapping, output_folder) print(f"Label mappings have been saved to {output_folder}")