diff --git a/tensnsorflow/applist.py b/tensnsorflow/applist.py index a101d162f069c8023c9672bb984732ba0979bff4..046de6b398f31d3ee4a538e3b636c268ca82203f 100644 --- a/tensnsorflow/applist.py +++ b/tensnsorflow/applist.py @@ -34,7 +34,7 @@ def sort_app(): df = df.rename(columns={0: "device_id", 1: "app_list"}) df = df.loc[df["app_list"].apply(is_json)] - category = {"competitor":{"新氧美容"}, + category = {"competitor":{"新氧美容","悦美","美呗整形","悦美微整形","如丽美容","医美咖","整形去哪儿","美黛拉","æ•´å½¢æ€å¯†è¾¾","美芽"}, "dianshang":{"京东","æ·˜å®","唯å“会","天猫","è‹å®æ˜“è´","国美","当当","亚马逊","网易严选","å°ç±³æœ‰å“"}, "kuajing_dianshang": {"å°çº¢ä¹¦", "网易考拉", "æ´‹ç 头", "达令全çƒå¥½è´§", "æµ·ç‹æµ·æ·˜", "HIG0", "豌豆公主", "å°šå“网", "丰趣海淘", "比呀比海外è´"}, @@ -72,7 +72,7 @@ def sort_app(): for i in category.keys(): df[i] = df["app_list"].apply(lambda x: 1 if len(x & category[i]) > 0 else 0) print(i) - print(df[i].unique()) + print(df[i].value_counts()) df = df.drop("app_list",axis=1) yconnect = create_engine('mysql+pymysql://root:3SYz54LS9#^9sBvC@10.66.157.22:4000/jerry_test?charset=utf8')