![](https://img-blog.csdnimg.cn/20201014180756928.png?x-oss-process=image/resize,m_fixed,h_64,w_64)
pandas
会spark的篮球少年
大数据分析,bi
展开
-
pandas自带图表
# 1.pandas 窗口函数,三角窗import pandas as pdimport numpy as np%matplotlib inlinedf = pd.DataFrame({'B': [0, 1, 2, 1, 4,2,3,0]})df['windows']=df.rolling(2, win_type='triang').sum()df.plot(style='o-',figsize=(10,5));2.核密度图df4 = pd.DataFrame({'a': np原创 2020-06-14 19:38:14 · 363 阅读 · 0 评论 -
python自带数据sqlite3学习,pandas to jeson
python自带数据sqlite3学习,pandas to jesonimport pandas as pdfrom sqlalchemy import create_engineengine = create_engine('sqlite:///foo.db', echo=True)# 建表#engine.execute("create table t1(id int,name...原创 2020-04-30 14:24:31 · 303 阅读 · 0 评论 -
pandas根据字段名炒粉excel
原创 2019-12-28 16:40:13 · 143 阅读 · 0 评论 -
pandas逆透视
#pandas 透视表 pivot_table'''The function pandas.pivot_table can be used to create spreadsheet-style pivot tables.It takes a number of argumentsdata: A DataFrame object values: a column or a ...原创 2019-01-07 15:13:10 · 1116 阅读 · 0 评论 -
group by cumsum sql ,pandas,saprk,tableau
Editgroup by cumsum sql ,pandas,saprk,tableau直接干不就完了,造数据: drop table if exists group_by_cumsum; create table if not exists group_by_cumsum(create_time datetime,city VarChar(20),sales...原创 2018-11-25 12:19:22 · 464 阅读 · 0 评论 -
tableau 上个月第一天,上个月今天
上个月今天:MAKEDATE(int(YEAR(TODAY())),month(DATEADD('month',-1,TODAY())),day(TODAY()))上个月第一天:MAKEDATE(int(YEAR(TODAY())),month(DATEADD('month',-1,TODAY())),1))原创 2018-04-02 16:57:09 · 3364 阅读 · 1 评论 -
pandas groupby year month
41down voteacceptedYou can use either resample or TimeGrouper (which resample uses under the hood).First make the datetime column is actually of datetimes (hit it with pd.to_datetime). It's easier if ...转载 2018-04-01 12:27:22 · 2267 阅读 · 0 评论 -
ValueError: cannot reindex a non-unique index with a method or limit
ValueError: cannot reindex a non-unique index with a method or limit是因为resample的2个date 相同 pandas 时间序列原创 2018-04-01 12:15:25 · 3775 阅读 · 0 评论 -
pandas时间序列
版权声明:本文为博主皮皮http://blog.csdn.net/pipisorry原创文章,未经博主允许不得转载。 https://blog.csdn.net/pipisorry/article/details/52209377目录(?)[+]http://blog.csdn.net/pipisorry/article/details/52209377pandas 最基本的时间序列类型就是以时间...转载 2018-04-01 10:37:28 · 308 阅读 · 0 评论 -
pandas to_period("M")
df['date2'] = df['服务截止日期'].dt.to_period("M")原创 2018-04-01 10:25:57 · 6969 阅读 · 4 评论