Skip to content

F
ffm-baseline
Commit 494f5ad5 authored by 张彦钊's avatar 张彦钊
Browse files
Options

change test file

parent 752ca9da
Hide whitespace changes
Inline Side-by-side
Showing with 21 additions and 21 deletions
tensnsorflow/feature_test.py
View file @ 494f5ad5
......@@ -151,7 +151,7 @@ def feature_engineer():
validate_date = con_sql(db, sql)[0].values.tolist()[0]
print("validate_date:" + validate_date)
temp = datetime.datetime.strptime(validate_date, "%Y-%m-%d")
start = (temp - datetime.timedelta(days=100)).strftime("%Y-%m-%d")
start = (temp - datetime.timedelta(days=3)).strftime("%Y-%m-%d")
print(start)
db = pymysql.connect(host='172.16.40.158', port=4000, user='root', passwd='3SYz54LS9#^9sBvC')
......@@ -214,7 +214,10 @@ def feature_engineer():
app_list_func(x[11], leve2_map), app_list_func(x[12], leve2_map),
[value_map[x[0]], value_map[x[13]], value_map[x[14]], value_map[x[15]], value_map[x[16]],
value_map[x[17]], value_map[x[18]], value_map[x[19]], value_map[x[20]], value_map[x[21]],
value_map[x[22]], value_map[x[23]], value_map[x[24]], value_map[x[25]], value_map[x[26]]]))
value_map[x[22]], value_map[x[23]], value_map[x[24]], value_map[x[25]], value_map[x[26]]]))\
.zipWithIndex().map(lambda x:(x[0][0],x[0][1],x[0][2],x[0][3],x[0][4],x[0][5],x[0][6],x[0][7],x[0][8],
x[0][9],x[0][10],x[0][11],x[0][12],x[0][13],
x[1]))
rdd.persist(storageLevel= StorageLevel.MEMORY_ONLY_SER)
......@@ -222,11 +225,11 @@ def feature_engineer():
train = rdd.map(
lambda x: (x[1], x[2], x[3], x[4], x[5], x[6], x[7], x[8], x[9],
x[10], x[11], x[12], x[13]))
x[10], x[11], x[12], x[13],x[14]))
f = time.time()
spark.createDataFrame(train).toDF("y", "z", "app_list", "level2_list", "level3_list",
"tag1_list", "tag2_list", "tag3_list", "tag4_list",
"tag5_list", "tag6_list", "tag7_list", "ids") \
"tag5_list", "tag6_list", "tag7_list", "ids","number") \
.repartition(1).write.format("tfrecords").save(path=path + "tr/", mode="overwrite")
h = time.time()
print("train tfrecord done")
......@@ -239,11 +242,11 @@ def feature_engineer():
test = rdd.filter(lambda x: x[0] == validate_date).map(
lambda x: (x[1], x[2], x[3], x[4], x[5], x[6], x[7], x[8], x[9],
x[10], x[11], x[12], x[13]))
x[10], x[11], x[12], x[13],x[14]))
spark.createDataFrame(test).toDF("y", "z", "app_list", "level2_list", "level3_list",
"tag1_list", "tag2_list", "tag3_list", "tag4_list",
"tag5_list", "tag6_list", "tag7_list", "ids") \
"tag5_list", "tag6_list", "tag7_list", "ids","number") \
.repartition(1).write.format("tfrecords").save(path=path + "va/", mode="overwrite")
print("va tfrecord done")
......@@ -302,18 +305,21 @@ def get_predict(date,value_map,app_list_map,leve2_map,leve3_map):
value_map.get(x[25], 11), value_map.get(x[26], 12),
value_map.get(x[27], 13), value_map.get(x[28], 14),
value_map.get(x[29], 15)
]))
]))\
.zipWithIndex().map(lambda x:(x[0][0],x[0][1],x[0][2],x[0][3],x[0][4],x[0][5],x[0][6],x[0][7],x[0][8],
x[0][9],x[0][10],x[0][11],x[0][12],x[0][13],x[0][14],x[0][15],x[0][16],
x[1]))
rdd.persist(storageLevel= StorageLevel.MEMORY_ONLY_SER)
native_pre = spark.createDataFrame(rdd.filter(lambda x:x[0] == 0).map(lambda x:(x[3],x[4],x[5])))\
.toDF("city","uid","cid_id")
print("native csv")
native_pre.repartition(1).write.format('com.databricks.spark.csv').save(path + "native/", header='true')
native_pre.toPandas().to_csv(local_path + "native.csv", header=True)
spark.createDataFrame(rdd.filter(lambda x: x[0] == 0)
.map(lambda x: (x[1],x[2],x[6],x[7],x[8],x[9],x[10],x[11],x[12],x[13],x[14],x[15],x[16]))) \
.map(lambda x: (x[1],x[2],x[6],x[7],x[8],x[9],x[10],x[11],x[12],x[13],x[14],x[15],x[16],x[17]))) \
.toDF("y","z","app_list", "level2_list", "level3_list","tag1_list", "tag2_list", "tag3_list", "tag4_list",
"tag5_list", "tag6_list", "tag7_list", "ids").repartition(1).write.format("tfrecords") \
"tag5_list", "tag6_list", "tag7_list", "ids","number").repartition(1).write.format("tfrecords") \
.save(path=path+"native/", mode="overwrite")
print("native tfrecord done")
h = time.time()
......@@ -322,13 +328,13 @@ def get_predict(date,value_map,app_list_map,leve2_map,leve3_map):
nearby_pre = spark.createDataFrame(rdd.filter(lambda x: x[0] == 1).map(lambda x: (x[3], x[4], x[5]))) \
.toDF("city", "uid", "cid_id")
print("nearby csv")
nearby_pre.repartition(1).write.format('com.databricks.spark.csv').save(path + "nearby/", header='true')
nearby_pre.toPandas().to_csv(local_path + "nearby.csv", header=True)
spark.createDataFrame(rdd.filter(lambda x: x[0] == 1)
.map(
lambda x: (x[1], x[2], x[6], x[7], x[8], x[9], x[10], x[11], x[12], x[13], x[14], x[15], x[16]))) \
lambda x: (x[1], x[2], x[6], x[7], x[8], x[9], x[10], x[11], x[12], x[13], x[14], x[15], x[16],x[17]))) \
.toDF("y", "z", "app_list", "level2_list", "level3_list", "tag1_list", "tag2_list", "tag3_list", "tag4_list",
"tag5_list", "tag6_list", "tag7_list", "ids").repartition(1).write.format("tfrecords") \
"tag5_list", "tag6_list", "tag7_list", "ids","number").repartition(1).write.format("tfrecords") \
.save(path=path + "nearby/", mode="overwrite")
print("nearby tfrecord done")
......@@ -349,14 +355,8 @@ if __name__ == '__main__':
spark.sparkContext.setLogLevel("WARN")
path = "hdfs:///strategy/esmm/"
local_path = "/home/gmuser/esmm/"
# validate_date, value_map, app_list_map, leve2_map, leve3_map = feature_engineer()
# get_predict(validate_date, value_map, app_list_map, leve2_map, leve3_map)
sql = "select y,ucity_id from jerry_test.esmm_train_data_dwell where stat_date >= '2019-06-11'"
df = spark.sql(sql).rdd.map(lambda x :(x[0],x[1])).zipWithIndex().map(lambda x:(x[0][0],x[0][1],x[1]))
spark.createDataFrame(df).show(6)
validate_date, value_map, app_list_map, leve2_map, leve3_map = feature_engineer()
get_predict(validate_date, value_map, app_list_map, leve2_map, leve3_map)
spark.stop()
......
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