python基础1

python基础

import os

import sys

import pandas as pd

from datetime import datetime, date

from time import time,localtime,strftime

import matplotlib.pylab as plt



dir_path = 'E:\\Data\\'

wdata = pd.read_csv(dir_path+'order_jx_y3',sep='\x01',encoding='utf-8',header=None)

wdata = wdata[[1,2,3,4,5]]

wdata.columns = ['province','city','sku','sku_qty','date']

wdata['date2'] = pd.to_datetime(wdata.date)

print(type(wdata['date2']),type(wdata['date']))

print(type(wdata.date.iloc[0]),type(wdata.date2.iloc[0]))

<class 'pandas.core.series.Series'> <class 'pandas.core.series.Series'>
<class 'str'> <class 'pandas._libs.tslib.Timestamp'>

wdata['day'] = pd.to_datetime(wdata.date2).dt.dayofweek

mask = (wdata['date2'] > '2016-12-31') & (wdata['date2']< '2018-01-01')

mask1=(wdata['day']==1)

g1 = wdata.loc[mask & mask1]

#也可以写成

mask = (wdata['date2'] > '2016-12-31') & (wdata['date2']< '2018-01-01')

&(wdata['day']==1)

g1 = wdata.loc[mask]

plot_data = g1[['sku_qty','date2']].groupby(['date2']).sum()

plot_data = plot_data.reset_index()

plot_data = g1[['province','sku_qty','date2']].groupby(['province','date2']).sum()

plot_data = plot_data.reset_index()

plot_data = plot_data.sort_values(by=['province','date2'])



#plot_data.head(5)

province_list = plot_data['province'].drop_duplicates().tolist()

for i, province in enumerate(province_list):

    mask_plot = plot_data['province'] == province

    plt.plot(plot_data['date2'].loc[mask_plot],plot_data['sku_qty'].loc[mask_plot])

#p = plot_data.groupby(['date2']).sum()

#plt.plot(p.index,p['sku_qty'])

plt.show()

for i, city in enumerate(city_list):

    mask_plot = plot_data['province'] == city

    plt.plot(plot_data['date2'].loc[mask_plot],plot_data['sku_qty'].loc[mask_plot])

p = plot_data.groupby(['date2']).sum()

plt.plot(p.index,p['sku_qty'])

plt.show()