6. RDD综合练习：更丰富的操作

学生课程分数

点击查看代码

from pyspark import SparkConf, SparkContext
conf = SparkConf().setMaster("local").setAppName("My App")
sc=SparkContext(conf=conf)

url = "file:///home/hadoop/input/sc.txt"

scm=sc.textFile(url).map(lambda line:line.split(',')).map(lambda line:[line[0],line[1],int(line[2])])

# 持久化

scm.cache()

# 总共有多少学生

scm.map(lambda a:a[0]).distinct().count()

# 开设了多少门课

scm.map(lambda a:a[1]).distinct().count()

# 转化为键值对

name = scm.map(lambda a:(a[0],(a[1],a[2])))

# 每个学生修了多少门课

name.countByKey()

# 每门课程有多少个学生选

name.values().countByKey()

# 有多少个100分

name.values().values().countByValue()[100]

# tom选了几门课？每门课多少分

print(len(name.lookup('Tom')))
name.lookup('Tom')

# tom选了几门课？每门课多少分

name.filter(lambda a:a[0]=='Tom').count()

name.filter(lambda a:a[0]=='Tom').collect()

# Tom的成绩按分数大小排序

name.filter(lambda a:a[0]=='Tom').sortBy(lambda a:a[1],False).collect()

# Tom的平均分

np.mean(name.filter(lambda a:a[0]=='Tom').values().values().collect())

# 每个分数+20平时分,并查看不及格人数的变化

name.filter(lambda a:a[1][1]<60).count()

notpass=name.map(lambda a:(a[0],(a[1][0],a[1][1]*0.8+20)))

notpass.filter(lambda a:a[1][1]<60).count()



notpass2=scm.map(lambda a:((a[0],a[1]),a[2])).mapValues(lambda v:v*0.8+20)

notpass2.filter(lambda a:a[1]<60).count()

# 求每门课的选修人数及平均分

course = sc.textFile(url).map(lambda a:a.split(',')).map(lambda a:(a[1],int(a[2])))

# 每门课的选修人数

countPeople=course.countByKey()
print(countPeople)

# lookup(),np.mean()实现

mean1={}
for i in countPeople.keys():
    mean1[i]=round(np.mean(course.lookup(i)),2)

# lookup(),np.mean()实现

mean1={}
for i in countPeople.keys():
    mean1[i]=round(np.mean(course.lookup(i)),2)

# lookup(),np.mean()实现

mean1={}
for i in countPeople.keys():
    mean1[i]=round(np.mean(course.lookup(i)),2)

print(mean1)

# reduceByKey()和collectAsMap()实现

totalCourse=course.reduceByKey(lambda a,b:a+b).collectAsMap()
totalCourse

mean2={}
for i in countPeople.keys():
    for j in totalCourse.keys():
        if(i==j):
            mean2[i]=round(totalCourse.get(i)/countPeople.get(i),2)

print(mean2)

# combineByKey()，map()，round()实现，确到2位小数

courseC=course.combineByKey(lambda a:(int(a),1),lambda a,b:(a[0]+int(b),a[1]+1),lambda a,b:(a[0]+b[0],a[1]+b[1]))
courseC.first()

courseC.map(lambda a:(a[0],a[1][1],round(a[1][0]/a[1][1],2))).collect()