From a9977262e4a51a5a76b0c26a085c74bc9b1f052b Mon Sep 17 00:00:00 2001
From: Haibin <1400012807@pku.edu.cn>
Date: Thu, 19 Dec 2024 16:14:19 +0800
Subject: [PATCH] meta tags aggregator

---
 data_juicer/ops/aggregator/meta_tags_aggregator.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/data_juicer/ops/aggregator/meta_tags_aggregator.py b/data_juicer/ops/aggregator/meta_tags_aggregator.py
index 454602657..a24025392 100644
--- a/data_juicer/ops/aggregator/meta_tags_aggregator.py
+++ b/data_juicer/ops/aggregator/meta_tags_aggregator.py
@@ -75,7 +75,7 @@ class MetaTagsAggregator(Aggregator):
                               '{tag_strs}')
     DEFAULT_TAG_TEMPLATE = '| {tag} | {cnt} |'
 
-    DEFAULT_OUTPUT_PATTERN = r'\n(.*?)归类为(.*?)(\Z|\n)'
+    DEFAULT_OUTPUT_PATTERN = r'\n(\w+)归类为(\w+)($|\n)'
 
     def __init__(self,
                  api_model: str = 'gpt-4o',