武西亮的博客

//如果这样每次重启的时候都会覆盖以前的日志，产生最新的日志。nohup bin/yanagishima-start.sh > y.log 2>&1 &//如果这样每次启动不会覆盖原来的老日志，日志会接着原来的日志继续写。nohup bin/yanagishima-start.sh >> y.log 2>&1 &...

经常遇到这种：'2021-01-27T09:38:40Z' UTC datetime格式，要转换成本地timezone格式 success_time = pd.Timestamp('2021-01-27T09:38:40Z'['completedAt'],tz="Asia/Shanghai").strftime("%Y-%m-%d %H:%M:%S") print(f"refresh done at {success_time}!")...

废话不说直接上代码：%matplotlib inlineimport numpy as npimport io,pandas as pdimport matplotlib.pyplot as pltmyString='''类别 订单量占比 门店数占比 交易额占比小龙虾 0.6 0.9 1.7川湘菜 4.5 4.8 6.5烤串 2.9 4.5 4.8面馆 3.8 8.0 3.1包子/粥 6.3 3.5 4.7西式快餐 7.9 7.5 9.2快餐简餐 22.1 16.3 18.8

mysql批量update死锁，因为提交太多，所以每隔100条commit一下就可以解决了，特此纪念！print("update started........")if df_new.shape[0]>0: for index,row in df_new.iterrows(): i += 1 print(i) #print(f"正在更新{row['shop_code']}") sql = f''' .

#connect to tableau serverimport tabpy_clientclient = tabpy_client.Client('http://localhost:9004/')#def function #deploy 函数到tableaudef clustering(x, y): import numpy as np from sklearn.cluster import DBSCAN from sklearn.preprocessing .

# 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

# 1.pandas 窗口函数，三角窗```pythonimport 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));```![6c7ee66f2d33a8e4

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...

# Linux查看并杀死被占用的端口sudo apt-get install lsof 　　//安装lsofsudo lsof -i:端口号　　　　　　//查找对应的进程号sudo kill -9 进程号　　　　　　//杀死对应的进程...

import concurrent.futuresimport urllib.requestimport pandas as pdimport numpy as npfrom sqlalchemy import create_engineimport pymysqlengine2 = create_engine('mysql+pymysql://root:wxl3322335@loc...

1.pycharm→preference→Editor→File and code Templates右侧找到Python Script,如下图2.配置文件如下：##!/Users/wxl/anaconda3/bin/python3.7# -*- coding: utf-8 -*-# @Time : ${DATE} ${TIME}# @Author : Mr w...

1) tableau正则REGEXP_EXTRACT(str([score]),"(\d{4}/\d{1,2}/\d{1,2})")REGEXP_EXTRACT(str([score]),"(\d+(\.\d+)?)”2) tableau正则提取字符前两位前几位：REGEXP_EXTRACT("人教版八年级语文abcd","版(.+年级)")REGEXP_EXTRACT("人...

https://blog.csdn.net/qq_33172735/article/details/87977583

生成器也是一种迭代器，但是你只能对其迭代一次。这是因为它们并没有把所有的值存在内存中，而是在运行时生成值。你通过遍历来使用它们，要么用一个“for”循环，要么将它们传递给任意可以进行迭代的函数和结构。大多数时候生成器是以函数来实现的。然而，它们并不返回一个值，而是yield(暂且译作“生出”)一个值。 一.python中的元组推导就是生成器：g=(x for x in range(5...

 #Python时间序列生成n=2dic=dict(zip([(pd.Timestamp("today")+ MonthBegin(n=-i-1)).strftime("%Y/%m/%d") for i in range(6)], [(pd.Timestamp("today")+ MonthBegin(n=-i)).strftime("%Y/%m/%d") fo...

 python鲜为人知的小技巧1.字典合并 以d2为基准更新d1d1={"a":1} d2={"b":2,"a":10} d1.update(d2) print(d1) print(dict(d1.items()|d2.items()))  2.for elsea=[1,2,3,4,5] for i in a: if i==0: print(i)...

直接上代码：粘贴格式很乱，可以直接点击我的印象笔记：https://app.yinxiang.com/shard/s60/nl/11100943/831e454b-b206-4b3f-93fb-00aa1e357172  1.建立spark appfrom pyspark.sql import SparkSessionbuilder = SparkSession.bu...

Skip to contentimpylaPython client for HiveServer2 implementations (e.g., Impala, Hive) for distributed query engines.For higher-level Impala functionality, including a Pandas-like interface over dist...

import pandas as pdimport numpy as npimport matplotlib.pyplot as pltfig, ax = plt.subplots()bar_positions=[1,2,3,4]bar_heights=[1,2,3,4]print(np.arange(len([2,2,3,4,5])+1))ax.bar(np.arange(len(...

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))

转(gedit:14333): Gtk-WARNING **: cannot open display: :0.02017年11月29日 09:28:45阅读数：983今天在终端下使用gedit时突然出现下面的错误：No protocol specified(gedit:14333): Gtk-WARNING **: cannot open display: :0.0开始还以为自己误操作将什么配置...

import browsercookie会出现以下报错：ModuleNotFoundError: No module named 'Crypto.Util.strxor'只要pip install pycryptodome就可以了

python -m pip install --upgrade pip升级pip的时候发现没有权限，这个时候只要用管理员权限打开命令行就好了! 

sudo ln -s /home/wxl/anaconda3/bin/python   /usr/bin/pythonln -s 就是新建连接，用前面的替代后面的路径ln -f 是覆盖修改，意思就是如果python已经被你天换成anaconda2了，这个时候你要变成anaconda3就需要用ln -f ln -s 新建连接ln -f 覆盖连接ln -i 删除连接...

ubuntu自带的python 版本是2.7，我们要把pyspark默认改成anaconda python 3.6down votYou can specify the version of Python for the driver by setting the appropriate environment variables in the ./conf/spark-env.sh ...

 第一步：生成config文件。终端输入：jupyter notebook --generate-config（如果是root用户请用：jupyter notebook --generate-config --allow-config）执行成功应该会显示：Writing default config to: /home/wxl/.jupyter/jupyter_notebo...

原文：https://blog.csdn.net/weixin_38168620/article/details/795965262018年03月17日 22:28:41 阅读数：2797 标签： 时间对象PandasTimestampPeriodTimedelta 更多个人分类： Python机器学习数据挖掘所属专栏： 数据分析利器--Pandas约定：import pan...

Python timedelta(转)2017年11月04日 21:59:15 阅读数：905更多个人分类： Python版权声明：本文为博主原创文章，未经博主允许不得转载。 https://blog.csdn.net/fengmm521/article/details/78446460datetime.timedelta对象代表两个时间之间的的时间差，两个date或datetim...

from pyspark.sql.column import Columnfrom pyspark.sql.column import _to_java_columnfrom pyspark.sql.column import _to_seq spark = SparkSession.builder.appName("scala_udf_test").getOrCreate()sc = ...

plotly可以制作交互式图表，直接上代码：import plotly.offline as pyfrom plotly.graph_objs import Scatter, Layoutimport plotly.graph_objs as gopy.init_notebook_mode(connected=True)import pandas as pdimport numpy...

import pandas as pdfrom matplotlib import pyplot as plt%matplotlib inlineimport seaborn as sns #读取数据df=pd.read_csv('seaborn.csv',sep=',',encoding='utf-8',index_col=0)df.head()  Name...

直接上代码: import pandas as pdimport numpy as npimport matplotlib.pyplot as pltfig, ax = plt.subplots()bar_positions=[1,2,3,4]bar_heights=[1,2,3,4]print(np.arange(len([2,2,3,4,5])+1))ax.bar(np.a...

python链接hive的各种疑难杂症1.首先安装impyla annoconda下执行以下命令：conda install -c anaconda impyla2.因为你hive肯定是linux，我这个是centos，所以当遇到c++问题就安装c++centos执行以下命令： yum install gcc-c++(centos中g++叫gcc-c++,如果直接安装g++会出现No packag...