Pandas数据处理

Pandas日期数据处理：如何按日期筛选、显示及统计数据

# -*- coding: utf-8 -*-
# @File : dataanalyse.py
# @Author : 段王爷
# @Time : 2021-11-25 20:11
# @Software: PyCharm
import datetime

import matplotlib
import matplotlib.pyplot as plt
import pandas as pd

matplotlib.rcParams['font.sans-serif'] = ['KaiTi', 'SimHei', 'FangSong']

x1 = datetime.datetime.strptime('2021-11-23', '%Y-%m-%d')
x2 = datetime.datetime.strptime('2021-11-21', '%Y-%m-%d')
x3 = datetime.datetime.strptime('2021-10-15', '%Y-%m-%d')
x4 = datetime.datetime.strptime('2021-09-15', '%Y-%m-%d')
a = [['A', '缺陷1', x1, x1], ['A', '缺陷2', x3, x3], ['A', '缺陷3', x2, x2],
     ['B', '缺陷4', x1, x1], ['B', '缺陷5', x1, x1], ['C', '缺陷6', x3, x3],
     ['A', '缺陷1', x1, x4], ['A', '缺陷1', x1, x4], ['C', '缺陷6', x3, x3]]
col = ['部门', '描述', '时间', '日期']
df = pd.Dataframe(a, columns=col)
df['时间'] = pd.to_datetime(df['时间'])
df.set_index('时间', inplace=True)
df = df.sort_index()
# 取某个月的数据
e = df['2021-10']
print(e)
# 按月统计
print(df.resample('M').count())

# 行数
num = df.shape[0]
print(num)
# dec = df.describe()
# print(dec)


# 最近一周
# r = df[df['时间'] > '2021-11-21']
# print(r)
# 按照部门和日期统计
print('*' * 100)
cf = df
# 将日期转成年月
cf['日期'] = cf.日期.apply(lambda x: x.strftime('%Y-%m')).astype('datetime64')  # 先按想要的格式转为字符串，再转为日期格式
print(cf)
r2 = cf.groupby([cf.部门, cf.日期]).count().unstack().fillna(0)
print(r2)
# 柱状图
# r2 = cf.groupby([cf.部门, cf.日期]).count().unstack().fillna(0).plot.bar()
# plt.show()
# 线图
r3 = cf.groupby([cf.部门, cf.日期]).日期.count().unstack().fillna(0)
# r4 = r3.plot.line()

# 单租数据绘图
# r5 = cf.groupby([cf.部门, cf.日期]).get_group(('A',))
# r5 = r3.loc['A'].plot.line
r3.loc['A'].plot()
r3.loc['B'].plot()
r3.loc['C'].plot()
# plt.xlabel('月份')
# plt.ylabel('bug数量')
# plt.legend
plt.show()
# r3 = df.groupby([df.部门, df['2021-11']]).count()
# print(r3)
# d = pd.pivot_table(df,index='部门',columns='时间',aggfunc='count')
# print(d)