处理用户模型

main_spider.py

@@ -26,11 +26,6 @@ class LightGBM(object):
         self.label_encoder = LabelEncoder()
         self.my_c = [
             "channel",
-            "view_count_user_30days",
-            "share_count_user_30days",
-            "return_count_user_30days",
-            "rov_user",
-            "str_user",
             "out_user_id",
             "mode",
             "out_play_cnt",
@@ -42,14 +37,9 @@ class LightGBM(object):
         ]
         self.str_columns = ["channel", "mode", "out_user_id", "tag1", "tag2", "tag3"]
         self.float_columns = [
-            "view_count_user_30days",
-            "share_count_user_30days",
-            "return_count_user_30days",
-            "rov_user",
-            "str_user",
             "out_play_cnt",
             "out_like_cnt",
-            "out_share_cnt",
+            "out_share_cnt"
         ]
         self.split_c = 0.7
         self.yc = 0.8

process_data.py

@@ -49,11 +49,6 @@ class DataProcessor(object):
         ]
         spider_features = [
             "channel",
-            "view_count_user_30days",
-            "share_count_user_30days",
-            "return_count_user_30days",
-            "rov_user",
-            "str_user",
             "out_user_id",
             "mode",
             "out_play_cnt",
@@ -154,10 +149,10 @@ if __name__ == "__main__":
     flag = int(input("please input method train or predict:\n "))
     if flag == 1:
         t = "train"
-        D = DataProcessor(flag=t, c="user")
+        D = DataProcessor(flag=t, c="spider")
         D.producer(dt="whole")
     else:
         t = "predict"
-        D = DataProcessor(flag=t, c="user")
+        D = DataProcessor(flag=t, c="spider")
         for d in range(16, 22):
             D.producer(d)