add gm_tag_cf

eda/smart_rank/gm_tag_cf.py

+import redis
+import json
 import numpy as np
 import pandas as pd
 from sklearn.metrics import pairwise_distances
@@ -49,20 +51,28 @@ if __name__ == '__main__':
     # 所有标签的协同标签
     all_tags_cf_tags = dict()
     for tag in all_tags_name:
-        if tag in cf_tags:
+        if tag in sorted_tag_ids:
             tag_cf_tags = get_similary_tags(tag, cf_tags, ratings_matrix)
         else:
-            tag_cf_tags = []
+            continue
         # 过滤父级标签
-        parent_tags = all_tag_parent_tags[tag]
+        parent_tags = all_tag_parent_tags.get(tag, [])  # 1级标签没有父级标签
         for parent_tag in parent_tags:
             if parent_tag in tag_cf_tags:
                 tag_cf_tags.remove(parent_tag)
         # 过滤标签的同义词
         tag_name = all_tags_name[tag]
         tag_cf_tags_names = [all_tags_name[i] for i in tag_cf_tags]
-        tag_synonym_names = all_word_synonym_words[tag_name]
+        tag_synonym_names = all_word_synonym_words.get(tag_name, [])  # 部分标签名称没有同义词
         for tag_synonym_name in tag_synonym_names:
             if tag_synonym_name in tag_cf_tags_names:
                 tag_cf_tags_names.remove(tag_synonym_name)
-        all_tags_cf_tags[tag_name] = tag_cf_tags_names
+        all_tags_cf_tags[tag_name] = json.dumps(tag_cf_tags_names[:10])
+
+    redis_client = redis.StrictRedis.from_url('redis://:ReDis!GmTx*0aN9@172.16.40.173:6379')
+    gm_tag_cf_tags_key = "gm:tag:cf:tags"
+    redis_client.hmset(gm_tag_cf_tags_key, all_tags_cf_tags)
+
+    # doris上执行
+    # data = redis_client.hgetall(gm_tag_cf_tags_key)
+    # data_dict = {str(i, 'utf-8'): json.loads(data[i]) for i in data}