diff --git a/.gitignore b/.gitignore
index 5781d3d..405ea37 100644
--- a/.gitignore
+++ b/.gitignore
@@ -50,7 +50,7 @@ images/nums/
 # 训练相关数据
 data_train/tmp/
 data_train/package/*.zip
-data_train/images/
+data_train/data_images/
 
 # 单独的截图文件
 screenshot.png
@@ -70,3 +70,4 @@ tmp
 # ===========================
 arknights.csv
 multi_ports.txt
+*.bak
diff --git a/auto_fetch.py b/auto_fetch.py
index 32f1717..0383d16 100644
--- a/auto_fetch.py
+++ b/auto_fetch.py
@@ -703,7 +703,7 @@ def auto_fetch_loop(self):
                     self._log(logging.INFO, "已达到设定时长，结束自动获取")
                     break
                 # 检测一次间隔时间——————————————————————————————————
-                time.sleep(0.2)
+                time.sleep(0.05)
             except Exception as e:
                 logger.exception(f"自动获取数据出错:\n{e}")
                 self._log(logging.ERROR, f"自动获取数据出错:\n{e}")
diff --git "a/data_train/\346\225\260\346\215\256\345\220\210\345\271\266.py" "b/data_train/\346\225\260\346\215\256\345\220\210\345\271\266.py"
index 4c31aa0..407870e 100644
--- "a/data_train/\346\225\260\346\215\256\345\220\210\345\271\266.py"
+++ "b/data_train/\346\225\260\346\215\256\345\220\210\345\271\266.py"
@@ -14,17 +14,7 @@
 if str(project_root) not in sys.path:
     sys.path.insert(0, str(project_root))
 
-#from config import MONSTER_COUNT, FIELD_FEATURE_COUNT config会读取图片导致非常慢
-
-def load_monster_data():
-    monster_data = pd.read_csv('monster_greenvine.csv', index_col="id", encoding='utf-8-sig')
-    return monster_data
-
-MONSTER_DATA = load_monster_data()
-
-# 全局变量
-MONSTER_COUNT = len(MONSTER_DATA)
-FIELD_FEATURE_COUNT = 0
+from config import MONSTER_COUNT, FIELD_FEATURE_COUNT # config会读取图片导致非常慢
 
 
 def get_expected_header():
@@ -65,7 +55,7 @@ def read_csv_from_zip(zip_ref, csv_filename):
 
 def process_archives(merge_images=True, extract_result_images=False):
     package_dir = base_dir / "package"
-    target_images_dir = base_dir / 'images'
+    target_images_dir = base_dir / 'data_images'
     target_csv_path = base_dir / 'arknights.csv'
 
     # 1. 目录准备 (移除旧的清空逻辑，保证增量更新)
@@ -137,7 +127,7 @@ def process_archives(merge_images=True, extract_result_images=False):
 
                     if header is None or header != expected_header:
                         print(f"  [跳过] {csv_member}: 表头不符合预期格式")
-                        continue
+                        # continue
 
                     zip_new_csv_rows = []
                     skip_csv_count = 0