Dataframe.plot调整图片大小的冲突问题

使用Dataframe.plot 时,无法使用 plt.figure 修改画布大小：

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

data=pd.read_csv('result.dat',sep='   ',engine='python',header=None)
data2=pd.read_csv('real.dat',sep=' ',engine='python',header=None)
data.columns = ['DORIS/RINEX 3.0','time(MJD)']
data2.columns = ['time(MJD)','doris 2.2']

result = pd.merge(data, data2, how='left', on='time(MJD)')
plt.figure(figsize=(20,10))

result.plot(x='time(MJD)',y=['DORIS/RINEX 3.0','doris 2.2'],kind='bar')

plt.locator_params(axis='x', nbins=10)
plt.autoscale()
plt.ylabel('observation numbers')
plt.tight_layout()
plt.savefig('ob.png',dpi=300)

可以看到应该是后者plot重新定义了新的画布，所以不能使用 matplotlib.plt.figure修改

直接使用Dataframe.plot()的参数修改即可 :

https://stackoverflow.com/questions/42215252/inconsistency-when-setting-figure-size-using-pandas-plot-method