小编dub*_*dan的帖子

我似乎可以弄清楚如何将句柄和标签传递matplotlib.patches.Patch给图例。

import matplotlib.pyplot as plt
import matplotlib.patches as mpatches


a_val = 0.6
colors = ['#EA5739','#FEFFBE','#4BB05C']


circ1 = mpatches.Patch( facecolor=colors[0],alpha=a_val,hatch=['\\\\'],label='Label1')
circ2= mpatches.Patch( facecolor=colors[1],alpha=a_val,hatch='o',label='Label2')
circ3 = mpatches.Patch(facecolor=colors[2],alpha=a_val,hatch='+',label='Label3')

fig,(ax) = plt.subplots()

ax.legend(handles = [circ1,circ2,circ3],loc=2)
plt.tight_layout()

为什么上例中的图例是空白的？

在数据框中，

 item_#, status, field1, field2 
   123,    "A",    "val1", "val2"
   223,    "B",    "val3", "val4"
   123,    "B",    "val5", "val6"
   323,    "A",    "val7", "val8"

item_#我想要的是同时具有 status"A"和 status的列表"B"。类似df.groupby('item_#')[(df.status.isin(['A', 'B']），但这实际上不起作用。它让我获得列表中任一值的所有项目。

任何建议将不胜感激！

我正在探索pandas.DataFrame.interpolate()不同的方法，linear与nearest，并且当尾部缺少数据时，我发现这两种方法的输出不同。

例如：

import pandas as pd # version: '0.16.2' or '0.20.3'
>>> a = pd.DataFrame({'col1': [np.nan, 1, np.nan, 3, np.nan, 5, np.nan]})
Out[1]: 
   col1
0   NaN
1   1.0
2   NaN
3   3.0
4   NaN
5   5.0
6   NaN

>>> a.interpolate(method='linear')
Out[2]: 
   col1
0   NaN
1   1.0
2   2.0
3   3.0
4   4.0
5   5.0
6   5.0

>>> a.interpolate(method='nearest')
Out[3]: 
   col1
0   NaN
1   1.0
2   1.0
3   3.0
4   3.0
5   5.0
6   NaN …

我有以下示例，其中我尝试向大于 999 的数字添加逗号分隔符（即 1,000 2,000 3,000 ...）

#Dummy data
Data1 <- data.frame(flow = c(8000,8.5,6,7.1,9), SP_elev = c(20,11,5,25,50))
Data2 <- data.frame(flow = c(7000,7.2,6.5,8.2,8.5), SP_elev = c(13,15,18,25,19))
Data3 <- data.frame(flow = c(2000,3,5,7,9), SP_elev = c(20,25,28,30,35))
Data4 <- data.frame(flow = c(1000,4,6,8,9), SP_elev = c(13,15,18,25,19))
Data5 <- data.frame(flow = c(1000,4,6,8,9), SP_elev = c(13,15,18,25,19))
Data6 <- data.frame(flow = c(1000,4,6,8,9), SP_elev = c(22,23,25,27,29))

#Create Vector list (in place of list.files)
dataframes = list("Data1" = Data1, 
                  "Data2" = Data2, 
                  "Data3" = Data3,
          "Data4" = Data4,
          "Data5" = …

我有一个数据帧query2:

    Site   TripDate       Volume
0   003l 1990-06-10  2202.571850
1   003l 1991-07-26  2543.566201
2   003l 1991-11-01  1702.228651
3   003l 1992-10-15  2753.163510
4   003l 1993-04-01  2550.538237
5   003l 1993-10-08  2241.329021

而另一个table1:

     TripDate  Count
0  1990-06-10     35
1  1991-07-26     35
2  1992-10-15     34
3  1993-10-08     35

我需要过滤query2,只包括TripDates在table1.生成的过滤表格如下所示:

    Site   TripDate       Volume
0   003l 1990-06-10  2202.571850
1   003l 1991-07-26  2543.566201
2   003l 1992-10-15  2753.163510
3   003l 1993-10-08  2241.329021

现在，当我的时间序列在十年之初（即 1990、2000、2010 年等）开始时，我有一些符合我的规范的工作代码，但我不知道如何调整我的代码以具有当我的时间序列从不是偶数的年份（即 1993 年）开始时的正确格式。

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

def format_xaxis(fig):

     years = dates.YearLocator(10,month=1,day=1)
     years1=dates.YearLocator(2,month=1,day=1)
     dfmt = dates.DateFormatter('%Y')
     dfmt1 = dates.DateFormatter('%y')

     [i.xaxis.set_major_locator(years) for i in fig.axes]
     [i.xaxis.set_minor_locator(years1) for i in fig.axes]
     [i.xaxis.set_major_formatter(dfmt) for i in fig.axes]
     [i.xaxis.set_minor_formatter(dfmt1) for i in fig.axes]
     [i.get_xaxis().set_tick_params(which='major', pad=15) for i in fig.axes]

     for t in fig.axes:
         for tick in t.xaxis.get_major_ticks():
             tick.label1.set_horizontalalignment('center')
         for label in t.get_xmajorticklabels() :
             label.set_rotation(0)
             label.set_weight('bold')
         for label in t.xaxis.get_minorticklabels():
             label.set_fontsize('small')
         for label in t.xaxis.get_minorticklabels()[::5]: …

我想从中获取结果pd.DataFrame.idxmax并使用它来更改索引前的值,并使用最大值.

如果我有df:

            Mule Creek  Saddle Mtn.  Calvert Creek
Date                                              
2011-05-01   23.400000    35.599998            8.6
2011-05-02   23.400000    35.599998            8.0
2011-05-03   23.400000    35.700001            7.6
2011-05-04   23.400000    50.000000            7.1
2011-05-05   23.100000    35.799999            6.4
2011-05-06   23.000000    35.799999            5.7
2011-05-07   40.000000    35.900002            4.7
2011-05-08   23.100000    36.500000           12.0
2011-05-09   23.299999    37.500000            4.4
2011-05-10   23.200001    37.500000            3.6

我发现每列的最大值出现在哪里:

max = df.idxmax()

我想确定最大值之前进行数值max全部np.nan

期望的结果:

            Mule Creek  Saddle Mtn.  Calvert Creek
Date                                              
2011-05-01         NaN          NaN            NaN
2011-05-02         NaN          NaN            NaN
2011-05-03         NaN          NaN …

我使用该pdf()设备创建了几个pdf ,我似乎无法从我的计算机中删除它们.我运行代码时收到以下错误.

> pdf(file="Appendix_B_1.pdf")
> plot(1:10)
> dev.off
function (which = dev.cur()) 
{
    if (which == 1) 
        stop("cannot shut down device 1 (the null device)")
    .External(C_devoff, as.integer(which))
    dev.cur()
}
<bytecode: 0x0000000008c7ecd0>
<environment: namespace:grDevices>

以上是我试图欺骗计算机认为它是一个新文件.它改变了文件,我可以打开它.我只是无法删除它.

我不知道这意味着什么,我想要做的就是删除文件.是否有一些解决方法可以关闭设备或从硬盘驱动器中删除文件？

我正在尝试将多个传递aggfuncs给pd.pivot_table：

如果我有new_df：

                 ATI     ATIMR
0 Basin Creek    2.0  0.039893
  Calvert Creek  0.0  0.006824
  Lick Creek     0.0  0.017371
  Mule Creek     0.0  0.041154
  Rocker Peak    2.0  0.027903
  Saddle Mtn.    0.0  0.052603
  Shower Falls   1.0  0.035456
1 Basin Creek    3.0  0.039893
  Calvert Creek  1.0  0.006824
  Lick Creek     1.0  0.017371

这有效：

   pct_75 = lambda y: np.percentile(y, 75)
   func_list = [np.median, np.mean ,pct_75]

   new_df = pd.pivot_table(new_df values='ATIMR',index='ATI',aggfunc=func_list)

但是当我尝试传递第二个lambda函数时，例如：

pct_25 = lambda x: np.percentile(x, 25)

func_list = [pct_25, …