longwei92的博客

import pandas as pdimport dateutil#获取当天的时间curr_time = pd.datetime.today()#对时间进行格式转换为yyyyMMddcurr_date = pd.datetime.strftime(curr_time, "%Y%m%d")#将日期减一天before_1_curr_time = curr_time - dateu...

创建数据集dataDict = {‘a’:[‘one’,‘two’], ‘b’:[[2,3],[1,3,6]], ‘c’:[1,2]}dataDf = pd.DataFrame(dataDict)实现行转列，drop(‘level_2’, axis = 1)：删掉新产生的索引，level_i中的i为groupby中列数df = t.groupby([‘a’, ‘c’]).b.apply(l...

地理编码工具geocoder地理编码工具geocoder，不同厂商的服务整合统一起来地理编码import geocoderg = geocoder.google("1403 Washington Ave, New Orleans, LA 70130")g = geocoder.arcgis(u"北京市海淀区上地十街10号")g.latlng输出为[29.9287839, -90....

import pandas as pddf = pd.DataFrame([['A',1,'2019-03-01 00:00:30'], ['A',3,'2019-04-01 00:00:30'], ['A',2,'2019-03-01 00:00:30'], ['B',5,'2019-...

lag该函数的格式如下:lag(字段名,N) over(partition by 分组字段 order by 排序字段 排序方式) lag括号里理由两个参数，第一个是字段名，第二个是数量N，这里的意思是，取分组排序后比该条记录序号小N的对应记录的指定字段的值，如果字段名为ts，N为1，就是取分组排序后上一条记录的ts值。lead该函数的格式如下:lead(字段名,N) over(pa...

import pandas as pddf = pd.read_table(read_file, header=0, dtype=str, sep='\t')res_file = fileDir + r'\filename.xlsx'writer = pd.ExcelWriter(res_file)groupby_df = df.groupby(['colname'])for i in...