本文整理汇总了Python中pysparkling.Context.map方法的典型用法代码示例。如果您正苦于以下问题：Python Context.map方法的具体用法？Python Context.map怎么用？Python Context.map使用的例子？那么恭喜您, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在类pysparkling.Context的用法示例。
在下文中一共展示了Context.map方法的9个代码示例，这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞，您的评价将有助于系统推荐出更棒的Python代码示例。
示例1: test_cache
# 需要导入模块: from pysparkling import Context [as 别名]
# 或者: from pysparkling.Context import map [as 别名]
def test_cache():
# this crashes in version 0.2.28
lines = Context().textFile('tests/*textFil*.py')
lines = lines.map(lambda l: '-'+l).cache()
print(len(lines.collect()))
lines = lines.map(lambda l: '+'+l)
lines = lines.map(lambda l: '-'+l).cache()
lines = lines.collect()
print(lines)
assert '-+-from pysparkling import Context' in lines示例2: test_lazy_execution_threadpool
# 需要导入模块: from pysparkling import Context [as 别名]
# 或者: from pysparkling.Context import map [as 别名]
def test_lazy_execution_threadpool():
def indent_line(l):
return '--- '+l
with futures.ThreadPoolExecutor(4) as p:
r = Context(pool=p).textFile('tests/test_multiprocessing.py')
r = r.map(indent_line).cache()
r.collect()
r = r.map(indent_line)
r = r.collect()
# ThreadPool is not lazy although it returns generators.
print(r)
assert '--- --- from pysparkling import Context' in r示例3: test_lazy_execution_processpool
# 需要导入模块: from pysparkling import Context [as 别名]
# 或者: from pysparkling.Context import map [as 别名]
def test_lazy_execution_processpool():
with futures.ProcessPoolExecutor(4) as p:
r = Context(
pool=p,
serializer=dill.dumps,
deserializer=dill.loads,
).textFile('tests/test_multiprocessing.py')
r = r.map(indent_line).cache()
r.collect()
r = r.map(indent_line)
r = r.collect()
# ProcessPool is not lazy although it returns generators.
print(r)
assert '--- --- from pysparkling import Context' in r示例4: test_lazy_execution
# 需要导入模块: from pysparkling import Context [as 别名]
# 或者: from pysparkling.Context import map [as 别名]
def test_lazy_execution():
r = Context().textFile('tests/test_multiprocessing.py')
r = r.map(indent_line)
exec_before_collect = INDENT_WAS_EXECUTED
# at this point, no map() or foreach() should have been executed
r.collect()
exec_after_collect = INDENT_WAS_EXECUTED
assert not exec_before_collect and exec_after_collect示例5: test_cache
# 需要导入模块: from pysparkling import Context [as 别名]
# 或者: from pysparkling.Context import map [as 别名]
def test_cache():
my_rdd = Context().parallelize([1, 2, 3, 4], 2)
my_rdd = my_rdd.map(lambda x: x*x).cache()
print('no exec until here')
print(my_rdd.first())
print('executed map on first partition only')
print(my_rdd.collect())
print('now map() was executed on all partitions and should '
'not be executed again')
print(my_rdd.collect())
assert len(my_rdd.collect()) == 4 and 16 in my_rdd.collect()示例6: test_lazy_execution_processpool
# 需要导入模块: from pysparkling import Context [as 别名]
# 或者: from pysparkling.Context import map [as 别名]
def test_lazy_execution_processpool():
def indent_line(l):
return '--- '+l
with futures.ProcessPoolExecutor(4) as p:
r = Context(
pool=p,
serializer=cloudpickle.dumps,
deserializer=pickle.loads,
).textFile('tests/test_multiprocessing.py') # .take(10)
print(r.collect())
r = r.map(indent_line)
print(r.collect())
r = r.cache()
print(r.collect())
r = r.map(indent_line)
r = r.collect()
# ProcessPool is not lazy although it returns generators.
print(r)
assert '--- --- from pysparkling import Context' in r示例7: test_processpool_distributed_cache
# 需要导入模块: from pysparkling import Context [as 别名]
# 或者: from pysparkling.Context import map [as 别名]
def test_processpool_distributed_cache():
with futures.ProcessPoolExecutor(4) as p:
r = Context(
pool=p,
serializer=cloudpickle.dumps,
deserializer=pickle.loads,
).parallelize(range(3), 3)
r = r.map(lambda _: time.sleep(0.1)).cache()
r.collect()
time_start = time.time()
print(r.collect())
time_end = time.time()
assert time_end - time_start < 0.3示例8: test_lazy_execution
# 需要导入模块: from pysparkling import Context [as 别名]
# 或者: from pysparkling.Context import map [as 别名]
def test_lazy_execution():
class I(object):
def __init__(self):
self.executed = False
def indent_line(self, l):
# global indent_was_executed
self.executed = True
return '--- '+l
r = Context().textFile('tests/test_multiprocessing.py')
i = I()
r = r.map(i.indent_line)
exec_before_collect = i.executed
# at this point, no map() or foreach() should have been executed
r = r.map(i.indent_line).cache()
print(r.collect())
r = r.map(i.indent_line)
r.collect()
exec_after_collect = i.executed
print((exec_before_collect, exec_after_collect))
assert not exec_before_collect and exec_after_collect示例9: test_concurrent
# 需要导入模块: from pysparkling import Context [as 别名]
# 或者: from pysparkling.Context import map [as 别名]
def test_concurrent():
with futures.ThreadPoolExecutor(4) as p:
my_rdd = Context(pool=p).parallelize([1, 3, 4])
r = my_rdd.map(math.sqrt).collect()
print(r)
assert 2 in r