pyspark-tutorial/tutorial/pyspark-examples/rdds/pyspark-session-2020-10-07.txt at master · mahmoudparsian/pyspark-tutorial · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
$ ./bin/pyspark
Python 3.7.2 (v3.7.2:9a3ffc0492, Dec 24 2018, 02:44:43)
[Clang 6.0 (clang-600.0.57)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
Using Spark's default log4j profile: org/apache/spark/log4j-defaults.properties
Setting default log level to "WARN".
To adjust logging level use sc.setLogLevel(newLevel). For SparkR, use setLogLevel(newLevel).
Welcome to
      ____              __
     / __/__  ___ _____/ /__
    _\ \/ _ \/ _ `/ __/  '_/
   /__ / .__/\_,_/_/ /_/\_\   version 3.0.0
      /_/

Using Python version 3.7.2 (v3.7.2:9a3ffc0492, Dec 24 2018 02:44:43)
SparkSession available as 'spark'.
>>> spark
<pyspark.sql.session.SparkSession object at 0x7fc7a913eba8>
>>>
>>>
>>>
>>> # create RDD[(String, Integer)]
...
>>> key_value_pairs =
[
 ('alex', 10),
 ('alex', 20),
 ('alex', 30),
 ('bob', 100),
 ('bob', 200),
 ('zazo', 7)
]

>>> # create an RDD[(String, Integer)] from a python collection
>>> key_value = spark.sparkContext.parallelize(key_value_pairs)
>>> key_value.count()
6
>>> key_value.collect()
[
 ('alex', 10),
 ('alex', 20),
 ('alex', 30),
 ('bob', 100),
 ('bob', 200),
 ('zazo', 7)
]
>>>
>>>
>>># use the reduceByKey() transformation
>>> sum_of_values_per_key = key_value.reduceByKey(lambda x, y: x+y)
>>>
>>> sum_of_values_per_key.count()
3
>>> sum_of_values_per_key.collect()
[
 ('bob', 300),
 ('alex', 60),
 ('zazo', 7)
]
>>>
>>>
>>>
>>> filtered = sum_of_values_per_key.filter(lambda x: x[1] > 10)
>>> filtered.collect()
[('bob', 300), ('alex', 60)]
>>>
>>>
>>> key_value.collect()
[
 ('alex', 10),
 ('alex', 20),
 ('alex', 30),
 ('bob', 100),
 ('bob', 200),
 ('zazo', 7)
]
>>>
>>> grouped = key_value.groupByKey()
>>> grouped.collect()
[
 ('bob', <pyspark.resultiterable.ResultIterable object at 0x7fc7a919f5c0>),
 ('alex', <pyspark.resultiterable.ResultIterable object at 0x7fc7a919f630>),
 ('zazo', <pyspark.resultiterable.ResultIterable object at 0x7fc7a919f588>)
]
>>> grouped.mapValues(lambda v : list(v)).collect()
[
 ('bob', [100, 200]),
 ('alex', [10, 20, 30]),
 ('zazo', [7])
]
>>> sum_of_values_per_key_2 = grouped.mapValues(lambda values: sum(values))
>>> sum_of_values_per_key_2.collect()
[
 ('bob', 300),
 ('alex', 60),
 ('zazo', 7)
]
>>>
>>>
>>> pairs = [('a', 10), ('a', 100), ('a', 200), ('b', 10)]
>>> rdd = spark.sparkContext.parallelize(pairs)
>>>
>>> rdd.collect()
[('a', 10), ('a', 100), ('a', 200), ('b', 10)]
>>> rdd2 = rdd.mapValues(lambda v: v+1000)
>>> rdd2.collect()
[('a', 1010), ('a', 1100), ('a', 1200), ('b', 1010)]
>>>
>>> rdd3 = rdd.map(lambda x: x[1]+1000)
>>> rdd3.collect()
[1010, 1100, 1200, 1010]
>>>
>>>
>>> rdd3 = rdd.map(lambda x: (x[0], x[1]+1000))
>>> rdd3.collect()
[('a', 1010), ('a', 1100), ('a', 1200), ('b', 1010)]
>>>
>>>
>>> data = [ ['a', 'b', 'c'], ['z'], [], [], ['alex', 'bob'] ]
>>> rdd = spark.sparkContext.parallelize(data)
>>> rdd.collect()
[['a', 'b', 'c'], ['z'], [], [], ['alex', 'bob']]
>>> rdd.count()
5
>>> flattened = rdd.flatMap(lambda x: x)
>>> flattened.count()
6
>>> flattened.collect()
['a', 'b', 'c', 'z', 'alex', 'bob']
>>> mapped = rdd.map(lambda x: x)
>>> mapped.count()
5
>>> mapped.collect()
[['a', 'b', 'c'], ['z'], [], [], ['alex', 'bob']]
>>>
>>>
>>> data = [ ['a', 'b', 'c'], ['z'], [], [], ('alex', 'bob') ]
>>> flattened2 = rdd.flatMap(lambda x: x)
>>> flattened2.collect()
['a', 'b', 'c', 'z', 'alex', 'bob']
>>>
>>>
>>>
>>> data2 = [ ['a', 'b', 'c'], ['z'], [], [], ('alex', 'bob') ]
>>> data2
[['a', 'b', 'c'], ['z'], [], [], ('alex', 'bob')]
>>> rdd2 = spark.sparkContext.parallelize(data2)
>>>
>>>
>>> rdd2.collect()
[['a', 'b', 'c'], ['z'], [], [], ('alex', 'bob')]
>>> rdd2.count()
5
>>> flattened2 = rdd2.flatMap(lambda x: x)
>>> flattened2.collect()
['a', 'b', 'c', 'z', 'alex', 'bob']
>>>
>>>
>>> data3 = [ ['a', 'b', 'c'], ['z'], [], [], 'alex', 'bob' ]
>>> rdd3 = spark.sparkContext.parallelize(data3)
>>> flattened3 = rdd3.flatMap(lambda x: x)
>>> flattened3.collect()
['a', 'b', 'c', 'z', 'a', 'l', 'e', 'x', 'b', 'o', 'b']
>>>