Skip to content

F
ffm-baseline
Commit 6bc26633 authored by 王志伟's avatar 王志伟
Browse files
Options

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

parents 47012369 9ca94cbe
Show whitespace changes
Inline Side-by-side
Showing with 40 additions and 41 deletions
eda/esmm/Feature_pipline/data2ffm.py
View file @ 6bc26633
...@@ -14,7 +14,7 @@ def con_sql(db,sql): ...@@ -14,7 +14,7 @@ def con_sql(db,sql):
try: try:
cursor.execute(sql) cursor.execute(sql)
result = cursor.fetchall() result = cursor.fetchall()
df = pd.DataFrame(list(result)).dropna() df = pd.DataFrame(list(result))
except Exception: except Exception:
print("发生异常", Exception) print("发生异常", Exception)
df = pd.DataFrame() df = pd.DataFrame()
...@@ -142,48 +142,47 @@ def get_data(): ...@@ -142,48 +142,47 @@ def get_data():
validate_date = con_sql(db, sql)[0].values.tolist()[0] validate_date = con_sql(db, sql)[0].values.tolist()[0]
print("validate_date:" + validate_date) print("validate_date:" + validate_date)
temp = datetime.datetime.strptime(validate_date, "%Y-%m-%d") temp = datetime.datetime.strptime(validate_date, "%Y-%m-%d")
start = (temp - datetime.timedelta(days=15)).strftime("%Y-%m-%d") start = (temp - datetime.timedelta(days=30)).strftime("%Y-%m-%d")
print(start) print(start)
db = pymysql.connect(host='10.66.157.22', port=4000, user='root', passwd='3SYz54LS9#^9sBvC', db='jerry_test') db = pymysql.connect(host='10.66.157.22', port=4000, user='root', passwd='3SYz54LS9#^9sBvC', db='jerry_test')
sql = "select e.device_id,e.y,e.z,e.stat_date,e.ucity_id,e.cid_id,e.clevel1_id,e.ccity_name," \ 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," \ "u.device_type,u.manufacturer,u.channel,c.top,cid_time.time " \
"home.jingxuan,home.zhibo,home.nose,home.eyes,home.weizheng,home.teeth,home.lunkuo," \
"home.meifu,home.xizhi,home.zhifang,home.longxiong,home.simi,home.maofa,home.gongli,home.korea " \
"from esmm_train_data e left join user_feature 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 home_tab_click home on e.device_id = home.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) "where e.stat_date >= '{}'".format(start)
df = con_sql(db, sql) df = con_sql(db, sql)
df = df.rename(columns={0: "device_id", 1: "y", 2: "z", 3: "stat_date", 4: "ucity_id", 5: "cid_id", print(df.shape)
6: "clevel1_id", 7: "ccity_name"}) 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"})
print("esmm data ok") print("esmm data ok")
print(df.head(2)) print(df.head(2))
ucity_id = list(set(df["ucity_id"].values.tolist()))
cid = list(set(df["cid_id"].values.tolist()))
df["clevel1_id"] = df["clevel1_id"].astype("str") df["clevel1_id"] = df["clevel1_id"].astype("str")
df["cid_id"] = df["cid_id"].astype("str")
df["y"] = df["y"].astype("str") df["y"] = df["y"].astype("str")
df["z"] = df["z"].astype("str") df["z"] = df["z"].astype("str")
df["y"] = df["stat_date"].str.cat([df["device_id"].values.tolist(),df["ucity_id"].values.tolist(), df["cid_id"].values.tolist(), df["top"] = df["top"].astype("str")
df["y"].values.tolist(),df["z"].values.tolist()], sep=",") df["y"] = df["stat_date"].str.cat([df["y"].values.tolist(),df["z"].values.tolist()], sep=",")
df = df.drop(["z","device_id"], axis=1).fillna(0.0) df = df.drop(["z","stat_date"], axis=1).fillna(0.0)
print(df.head(2)) print(df.head(2))
features = 0
for i in ["ucity_id","clevel1_id","ccity_name","device_type","manufacturer","channel"]:
features = features + len(df[i].unique())
print("fields:{}".format(df.shape[1]-1)) print("fields:{}".format(df.shape[1]-1))
print("features:{}".format(len(cid))) print("features:{}".format(features))
return df,validate_date,ucity_id,cid ccity_name = list(set(df["ccity_name"].values.tolist()))
ucity_id = list(set(df["ucity_id"].values.tolist()))
return df,validate_date,ucity_id,ccity_name
def transform(a,validate_date): def transform(a,validate_date):
model = multiFFMFormatPandas() model = multiFFMFormatPandas()
df = model.fit_transform(a, y="y", n=160000, processes=26) df = model.fit_transform(a, y="y", n=160000, processes=22)
df = pd.DataFrame(df) df = pd.DataFrame(df)
df["stat_date"] = df[0].apply(lambda x: x.split(",")[0]) df["stat_date"] = df[0].apply(lambda x: x.split(",")[0])
df["device_id"] = df[0].apply(lambda x: x.split(",")[1])
df["city_id"] = df[0].apply(lambda x: x.split(",")[2])
df["cid"] = df[0].apply(lambda x: x.split(",")[3])
df["number"] = np.random.randint(1, 2147483647, df.shape[0]) df["number"] = np.random.randint(1, 2147483647, df.shape[0])
df["seq"] = list(range(df.shape[0])) df["seq"] = list(range(df.shape[0]))
df["seq"] = df["seq"].astype("str") df["seq"] = df["seq"].astype("str")
df["data"] = df[0].apply(lambda x: ",".join(x.split(",")[4:])) df["data"] = df[0].apply(lambda x: ",".join(x.split(",")[1:]))
df["data"] = df["seq"].str.cat(df["data"], sep=",") df["data"] = df["seq"].str.cat(df["data"], sep=",")
df = df.drop([0,"seq"], axis=1) df = df.drop([0,"seq"], axis=1)
print(df.head(2)) print(df.head(2))
...@@ -192,42 +191,42 @@ def transform(a,validate_date): ...@@ -192,42 +191,42 @@ def transform(a,validate_date):
train = train.drop("stat_date",axis=1) train = train.drop("stat_date",axis=1)
test = df[df["stat_date"] == validate_date] test = df[df["stat_date"] == validate_date]
test = test.drop("stat_date",axis=1) test = test.drop("stat_date",axis=1)
print("train shape") # print("train shape")
print(train.shape) # print(train.shape)
train.to_csv(path + "tr.csv", sep="\t", index=False) train.to_csv(path + "tr.csv", sep="\t", index=False)
test.to_csv(path + "va.csv", sep="\t", index=False) test.to_csv(path + "va.csv", sep="\t", index=False)
return model return model
def get_predict_set(ucity_id, cid,model): def get_predict_set(ucity_id,model,ccity_name):
db = pymysql.connect(host='10.66.157.22', port=4000, user='root', passwd='3SYz54LS9#^9sBvC', db='jerry_test') db = pymysql.connect(host='10.66.157.22', port=4000, user='root', passwd='3SYz54LS9#^9sBvC', db='jerry_test')
sql = "select e.device_id,e.y,e.z,e.stat_date,e.ucity_id,e.cid_id,e.clevel1_id,e.ccity_name," \ sql = "select e.y,e.z,e.label,e.ucity_id,e.clevel1_id,e.ccity_name," \
"u.device_type,u.manufacturer,u.channel," \ "u.device_type,u.manufacturer,u.channel,c.top,cid_time.time,e.device_id,e.cid_id " \
"home.jingxuan,home.zhibo,home.nose,home.eyes,home.weizheng,home.teeth,home.lunkuo," \
"home.meifu,home.xizhi,home.zhifang,home.longxiong,home.simi,home.maofa,home.gongli,home.korea,e.label " \
"from esmm_pre_data e left join user_feature 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 home_tab_click home on e.device_id = home.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 = con_sql(db, sql)
df = df.rename(columns={0: "device_id", 1: "y", 2: "z", 3: "stat_date", 4: "ucity_id", 5: "cid_id", df = df.rename(columns={0: "y", 1: "z", 2: "label", 3: "ucity_id", 4: "clevel1_id", 5: "ccity_name",
6: "clevel1_id", 7: "ccity_name",26:"label"}) 6: "device_type", 7: "manufacturer", 8: "channel", 9: "top", 10: "time",
11:"device_id",12:"cid_id"})
print("before filter:") print("before filter:")
print(df.shape) print(df.shape)
df = df[df["cid_id"].isin(cid)]
print("after cid filter:")
print(df.shape)
df = df[df["ucity_id"].isin(ucity_id)] df = df[df["ucity_id"].isin(ucity_id)]
print("after ucity filter:") print("after ucity filter:")
print(df.shape) print(df.shape)
df["clevel1_id"] = df["clevel1_id"].astype("str") df = df[df["ccity_name"].isin(ccity_name)]
print("after ccity_name filter:")
print(df.shape)
df["cid_id"] = df["cid_id"].astype("str") df["cid_id"] = df["cid_id"].astype("str")
df["clevel1_id"] = df["clevel1_id"].astype("str")
df["top"] = df["top"].astype("str")
df["y"] = df["y"].astype("str") df["y"] = df["y"].astype("str")
df["z"] = df["z"].astype("str") df["z"] = df["z"].astype("str")
df["label"] = df["label"].astype("str") df["label"] = df["label"].astype("str")
df["y"] = df["label"].str.cat( df["y"] = df["label"].str.cat(
[df["device_id"].values.tolist(), df["ucity_id"].values.tolist(), df["cid_id"].values.tolist(), [df["device_id"].values.tolist(), df["ucity_id"].values.tolist(), df["cid_id"].values.tolist(),
df["y"].values.tolist(), df["z"].values.tolist()], sep=",") df["y"].values.tolist(), df["z"].values.tolist()], sep=",")
df = df.drop(["z","label","device_id"], axis=1).fillna(0.0) df = df.drop(["z","label","device_id","cid_id"], axis=1).fillna(0.0)
print(df.head(2)) print(df.head(2))
df = model.transform(df,n=160000, processes=22) df = model.transform(df,n=160000, processes=22)
df = pd.DataFrame(df) df = pd.DataFrame(df)
...@@ -260,9 +259,9 @@ def get_predict_set(ucity_id, cid,model): ...@@ -260,9 +259,9 @@ def get_predict_set(ucity_id, cid,model):
if __name__ == "__main__": if __name__ == "__main__":
path = "/home/gaoyazhe/data/" path = "/home/gaoyazhe/data/"
a = time.time() a = time.time()
df, validate_date, ucity_id, cid = get_data() df, validate_date, ucity_id,ccity_name = get_data()
model = transform(df, validate_date) model = transform(df, validate_date)
get_predict_set(ucity_id, cid,model) get_predict_set(ucity_id,model,ccity_name)
b = time.time() b = time.time()
print("cost(分钟)") print("cost(分钟)")
print((b-a)/60) print((b-a)/60)
eda/esmm/Feature_pipline/get_tfrecord.py
View file @ 6bc26633
...@@ -25,8 +25,8 @@ tf.app.flags.DEFINE_integer("threads", 16, "threads num") ...@@ -25,8 +25,8 @@ tf.app.flags.DEFINE_integer("threads", 16, "threads num")
#User_Fileds = set(['101','109_14','110_14','127_14','150_14','121','122','124','125','126','127','128','129']) #User_Fileds = set(['101','109_14','110_14','127_14','150_14','121','122','124','125','126','127','128','129'])
#Ad_Fileds = set(['205','206','207','210','216']) #Ad_Fileds = set(['205','206','207','210','216'])
#Context_Fileds = set(['508','509','702','853','301']) #Context_Fileds = set(['508','509','702','853','301'])
Common_Fileds = {'1':'1','2':'2','3':'3','4':'4','5':'5','6':'6','7':'7','8':'8','9':'9','10':'10','11':'11','12':'12','13':'13','14':'14','15':'15','16':'16','17':'17','18':'18','19':'19','20':'20','21':'21','22':'22','23':'23'} #Common_Fileds = {'1':'1','2':'2','3':'3','4':'4','5':'5','6':'6','7':'7','8':'8','9':'9','10':'10','11':'11','12':'12','13':'13','14':'14','15':'15','16':'16','17':'17','18':'18','19':'19','20':'20','21':'21','22':'22','23':'23'}
#Common_Fileds = {'1':'1','2':'2','3':'3','4':'4','5':'5','6':'6','7':'7','8':'8','9':'9','10':'10','11':'11'} Common_Fileds = {'1':'1','2':'2','3':'3','4':'4','5':'5','6':'6','7':'7','8':'8'}
UMH_Fileds = {'109_14':('u_cat','12'),'110_14':('u_shop','13'),'127_14':('u_brand','14'),'150_14':('u_int','15')} #user multi-hot feature UMH_Fileds = {'109_14':('u_cat','12'),'110_14':('u_shop','13'),'127_14':('u_brand','14'),'150_14':('u_int','15')} #user multi-hot feature
Ad_Fileds = {'206':('a_cat','16'),'207':('a_shop','17'),'210':('a_int','18'),'216':('a_brand','19')} #ad feature for DIN Ad_Fileds = {'206':('a_cat','16'),'207':('a_shop','17'),'210':('a_int','18'),'216':('a_brand','19')} #ad feature for DIN
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment