创建 PySpark 数据框:年份的月份序列 [英] Create PySpark dataframe : sequence of months with year

问题描述

这里完全是新手.

我想使用 pyspark 创建一个 dataframe，它将列出月份和年份，采用当前日期并列出 x 行.

I would like to create a dataframe using pyspark that will list month and year taking the current date and listing x number of lines.

如果我决定 x=5 数据帧应该如下所示

if i decide x=5 dataframe should like as below

Calendar_Entry

August 2019<br/>
September 2019<br/>
October 2019<br/>
November 2019<br/>
December 2019

推荐答案

Spark 不是以分布式方式生成行的工具，而是用于处理然后分布式的工具.
由于无论如何您的数据都很小，因此最好的解决方案可能是在纯 python 中创建数据，并在需要时从中创建一个 spark 数据框.

Spark is not a tool for generating rows in a distributed way but rather for processing then distributed.
Since your data is small anyway the best solution is probably to create the data in pure python and if required create a spark dataframe out of it.

import datetime
from dateutil.relativedelta import relativedelta


def create_months_df(n_months):
    date_list = [datetime.datetime.today() - relativedelta(months=i) for i in range(n_months)]
    dates_formatted = [(d.strftime("%B"), d.year) for d in date_list]
    return spark.createDataFrame(dates_formatted, ["month", "year"])

这篇关于创建 PySpark 数据框:年份的月份序列的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！