增加打印

eda/esmm/Model_pipline/feature_engineering.py

@@ -111,27 +111,27 @@ def feature_engineer():
     unique_values.extend(get_unique(db, sql))
 
     db = pymysql.connect(host='172.16.40.158', port=4000, user='root', passwd='3SYz54LS9#^9sBvC', db='jerry_test')
-    sql = "select distinct price_min from train_Knowledge_network_data"
+    sql = "select distinct price_min from knowledge"
     unique_values.extend(get_unique(db, sql))
 
     db = pymysql.connect(host='172.16.40.158', port=4000, user='root', passwd='3SYz54LS9#^9sBvC', db='jerry_test')
-    sql = "select distinct treatment_method from train_Knowledge_network_data"
+    sql = "select distinct treatment_method from knowledge"
     unique_values.extend(get_unique(db, sql))
 
     db = pymysql.connect(host='172.16.40.158', port=4000, user='root', passwd='3SYz54LS9#^9sBvC', db='jerry_test')
-    sql = "select distinct price_max from train_Knowledge_network_data"
+    sql = "select distinct price_max from knowledge"
     unique_values.extend(get_unique(db, sql))
 
     db = pymysql.connect(host='172.16.40.158', port=4000, user='root', passwd='3SYz54LS9#^9sBvC', db='jerry_test')
-    sql = "select distinct treatment_time from train_Knowledge_network_data"
+    sql = "select distinct treatment_time from knowledge"
     unique_values.extend(get_unique(db, sql))
 
     db = pymysql.connect(host='172.16.40.158', port=4000, user='root', passwd='3SYz54LS9#^9sBvC', db='jerry_test')
-    sql = "select distinct maintain_time from train_Knowledge_network_data"
+    sql = "select distinct maintain_time from knowledge"
     unique_values.extend(get_unique(db, sql))
 
     db = pymysql.connect(host='172.16.40.158', port=4000, user='root', passwd='3SYz54LS9#^9sBvC', db='jerry_test')
-    sql = "select distinct recover_time from train_Knowledge_network_data"
+    sql = "select distinct recover_time from knowledge"
     unique_values.extend(get_unique(db, sql))
 
     db = pymysql.connect(host='172.16.40.158', port=4000, user='root', passwd='3SYz54LS9#^9sBvC', db='jerry_test')
@@ -154,6 +154,8 @@ def feature_engineer():
     unique_values.extend(features)
     print("unique_values length")
     print(len(unique_values))
+    print("特征维度：")
+    print(apps_number + level2_number + level3_number + len(unique_values))
 
     temp = list(range(16 + apps_number + level2_number + level3_number,
                       16 + apps_number + level2_number + level3_number + len(unique_values)))
@@ -169,7 +171,7 @@ def feature_engineer():
           "left join jerry_test.cid_time_cut cut on e.cid_id = cut.cid " \
           "left join jerry_test.device_app_list dl on e.device_id = dl.device_id " \
           "left join jerry_test.diary_feat feat on e.cid_id = feat.diary_id " \
-          "left join jerry_test.train_Knowledge_network_data k on feat.level2 = k.level2_id " \
+          "left join jerry_test.knowledge k on feat.level2 = k.level2_id " \
           "left join jerry_test.wiki_tag wiki on e.device_id = wiki.device_id " \
           "left join jerry_test.question_tag question on e.device_id = question.device_id " \
           "left join jerry_test.search_tag search on e.device_id = search.device_id " \
@@ -254,7 +256,7 @@ def get_predict(date,value_map,app_list_map,leve2_map,leve3_map):
           "left join jerry_test.order_tag ot on e.device_id = ot.device_id " \
           "left join jerry_test.sixin_tag sixin on e.device_id = sixin.device_id " \
           "left join jerry_test.cart_tag cart on e.device_id = cart.device_id " \
-          "left join jerry_test.train_Knowledge_network_data k on feat.level2 = k.level2_id"
+          "left join jerry_test.knowledge k on feat.level2 = k.level2_id"
 
     features = ["ucity_id", "ccity_name", "device_type", "manufacturer",
                 "channel", "top", "time", "hospital_id",