pyspark运行ALS推荐算法

首先创建读取路径，这个取决于你的pyspark是以什么方式启动的

global Path
if sc.master[0:5] == 'local':
    Path='file:/home/swt/pythonwork/PythonProject/'
else:
    Path="hdfs://ubuntu:9000/user/swt/"

我是本地启动

sc.master
'local[*]'

开始读取数据

rawUserData = sc.textFile(Path+"data/u.data")
rawUserData.count()
rawUserData.first()
rawRatings = rawUserData.map(lambda line:line.split("\t")[:3])
rawRatings.take(5)
ratingsRDD = rawRatings.map(lambda x:(x[0],x[1],x[2]))
ratingsRDD.take(5)

100000
'196\t242\t3\t881250949'
[['196', '242', '3'],
 ['186', '302', '3'],
 ['22', '377', '1'],
 ['244', '51', '2'],
 ['166', '346', '1']]
[('196', '242', '3'),
 ('186', '302', '3'),
 ('22', '377', '1'),
 ('244', '51', '2'),
 ('166', '346', '1')]

# 对数据进行处理
numRatings=ratingsRDD.count()
numRatings
numUsers = ratingsRDD.map(lambda x:x[0]).distinct().count()
numUsers
numMovles = ratingsRDD.map(lambda x:x[1]).distinct().count()
numMovles

100000
943
1682

# 将处理好的数据进行训练，ALS算法返回的是一个model
from pyspark.mllib.recommendation import ALS
model = ALS.train(ratingsRDD, 10, 10, 0.01)
print(model)

<pyspark.mllib.recommendation.MatrixFactorizationModel object at 0x7fb87179b908>

# 用户id为100的用户推荐以下电影和推荐指数
model.recommendProducts(100,5)

[Rating(user=100, product=1160, rating