Merge branch 'master' of http://git.wanmeizhensuo.com/ML/ffm-baseline

eda/esmm/Feature_pipline/data2ffm.py

@@ -147,7 +147,7 @@ def get_data():
     db = pymysql.connect(host='10.66.157.22', port=4000, user='root', passwd='3SYz54LS9#^9sBvC', db='jerry_test')
     sql = "select e.y,e.z,e.stat_date,e.ucity_id,e.clevel1_id,e.ccity_name," \
           "u.device_type,u.manufacturer,u.channel,c.top,cid_time.time,e.device_id " \
-          "from esmm_train_data e left join user_feature_clean u on e.device_id = u.device_id " \
+          "from esmm_train_data e left join user_feature u on e.device_id = u.device_id " \
           "left join cid_type_top c on e.device_id = c.device_id left join cid_time on e.cid_id = cid_time.cid_id " \
           "where e.stat_date >= '{}'".format(start)
     df = con_sql(db, sql)
@@ -208,7 +208,7 @@ def get_predict_set(ucity_id,model,ccity_name,manufacturer,channel):
     db = pymysql.connect(host='10.66.157.22', port=4000, user='root', passwd='3SYz54LS9#^9sBvC', db='jerry_test')
     sql = "select e.y,e.z,e.label,e.ucity_id,e.clevel1_id,e.ccity_name," \
           "u.device_type,u.manufacturer,u.channel,c.top,cid_time.time,e.device_id,e.cid_id " \
-          "from esmm_pre_data e left join user_feature_clean u on e.device_id = u.device_id " \
+          "from esmm_pre_data e left join user_feature u on e.device_id = u.device_id " \
           "left join cid_type_top c on e.device_id = c.device_id left join cid_time on e.cid_id = cid_time.cid_id"
     df = con_sql(db, sql)
     df = df.rename(columns={0: "y", 1: "z", 2: "label", 3: "ucity_id", 4: "clevel1_id", 5: "ccity_name",

tensnsorflow/ffm.py

@@ -145,55 +145,35 @@ def get_data():
     temp = datetime.datetime.strptime(validate_date, "%Y-%m-%d")
     start = (temp - datetime.timedelta(days=30)).strftime("%Y-%m-%d")
     print(start)
-    db = pymysql.connect(host='10.66.157.22', port=4000, user='root', passwd='3SYz54LS9#^9sBvC')
-    # sql = "select e.y,e.z,e.stat_date,e.ucity_id,e.clevel1_id,e.ccity_name," \
-    #       "u.device_type,u.manufacturer,u.channel,c.top,cid_time.time,e.device_id," \
-    #       "a.competitor,a.dianshang,a.kuajing_dianshang,a.zhibo,a.youxizhibo,a.short_video,a.meitu,a.tiyu," \
-    #       "a.read,a.finance,a.fashion_clothes,a.muying,a.fresh,a.bijia,a.travel,a.airplane," \
-    #       "a.love,a.stock,a.car,a.child,a.homework,a.work,a.job " \
-    #       "from esmm_train_data e left join user_feature_clean u on e.device_id = u.device_id " \
-    #       "left join cid_type_top c on e.device_id = c.device_id " \
-    #       "left join cid_time on e.cid_id = cid_time.cid_id " \
-    #       "left join app_list_sort a on e.device_id = a.device_id " \
-    #       "where e.stat_date >= '{}'".format(start)
-
+    db = pymysql.connect(host='10.66.157.22', port=4000, user='root', passwd='3SYz54LS9#^9sBvC', db='jerry_test')
     sql = "select e.y,e.z,e.stat_date,e.ucity_id,e.clevel1_id,e.ccity_name," \
-          "u.device_type,u.manufacturer,u.channel,c.top,cid_time.time,e.device_id,feat.max_level1_id " \
-          "from jerry_test.esmm_train_data e " \
-          "left join jerry_test.user_feature_clean u on e.device_id = u.device_id " \
-          "left join jerry_test.cid_type_top c on e.device_id = c.device_id " \
-          "left join jerry_test.cid_time on e.cid_id = cid_time.cid_id " \
-          "left join jerry_prod.device_feat feat on e.device_id = feat.device_id " \
+          "u.device_type,u.manufacturer,u.channel,c.top,cid_time.time,e.device_id " \
+          "from esmm_train_data e left join user_feature u on e.device_id = u.device_id " \
+          "left join cid_type_top c on e.device_id = c.device_id left join cid_time on e.cid_id = cid_time.cid_id " \
           "where e.stat_date >= '{}'".format(start)
     df = con_sql(db, sql)
     print(df.shape)
-    df = df.rename(columns={0: "y", 1: "z", 2: "stat_date", 3: "ucity_id", 4: "clevel1_id", 5: "ccity_name",
-                            6: "device_type", 7: "manufacturer", 8: "channel", 9: "top", 10: "time",
-                            11: "device_id",12:"max_level1_id"})
-
+    df = df.rename(columns={0: "y", 1: "z", 2: "stat_date", 3: "ucity_id",4: "clevel1_id", 5: "ccity_name",
+                            6:"device_type",7:"manufacturer",8:"channel",9:"top",10:"time",11:"device_id"})
     print("esmm data ok")
     print(df.head(2))
-    print(df.count())
 
     df["clevel1_id"] = df["clevel1_id"].astype("str")
     df["y"] = df["y"].astype("str")
     df["z"] = df["z"].astype("str")
     df["top"] = df["top"].astype("str")
     df["y"] = df["stat_date"].str.cat([df["device_id"].values.tolist(),df["y"].values.tolist(),df["z"].values.tolist()], sep=",")
-    df = df.drop(["z","stat_date","device_id"], axis=1)
+    df = df.drop(["z","stat_date","device_id"], axis=1).fillna(0.0)
+    print(df.head(2))
     features = 0
-    for i in ["ucity_id","clevel1_id","ccity_name","device_type","manufacturer","channel","top","max_level1_id"]:
+    for i in ["ucity_id","clevel1_id","ccity_name","device_type","manufacturer","channel"]:
         features = features + len(df[i].unique())
-        df[i] = df[i].fillna("a")
-    df["time"] = df["time"].fillna(df["time"].mean())
-    print(df.head(2))
     print("fields:{}".format(df.shape[1]-1))
     print("features:{}".format(features))
     ccity_name = list(set(df["ccity_name"].values.tolist()))
     ucity_id = list(set(df["ucity_id"].values.tolist()))
     manufacturer = list(set(df["manufacturer"].values.tolist()))
     channel = list(set(df["channel"].values.tolist()))
-
     return df,validate_date,ucity_id,ccity_name,manufacturer,channel
 
 
@@ -229,7 +209,7 @@ def get_predict_set(ucity_id,model,ccity_name,manufacturer,channel):
     db = pymysql.connect(host='10.66.157.22', port=4000, user='root', passwd='3SYz54LS9#^9sBvC', db='jerry_test')
     sql = "select e.y,e.z,e.label,e.ucity_id,e.clevel1_id,e.ccity_name," \
           "u.device_type,u.manufacturer,u.channel,c.top,cid_time.time,e.device_id,e.cid_id " \
-          "from esmm_pre_data e left join user_feature_clean u on e.device_id = u.device_id " \
+          "from esmm_pre_data e left join user_feature u on e.device_id = u.device_id " \
           "left join cid_type_top c on e.device_id = c.device_id left join cid_time on e.cid_id = cid_time.cid_id"
     df = con_sql(db, sql)
     df = df.rename(columns={0: "y", 1: "z", 2: "label", 3: "ucity_id", 4: "clevel1_id", 5: "ccity_name",