Pandas read#U html正在创建一个具有2x'；列数是多少

我创建了一个数据框架，从一个站点获取一些数据

df = pd.read_html('http://finviz.com/insidertrading.ashx?tc=1', header = 0)[4].set_index('Date')

然后，我创建了一个html文件，以html文件的名称作为日期

     today_date = dt.date.today().isoformat()
    html_name = 'Insider Trading/{}_buys.html'.format(today_date)
    df.to_html(html_name)

当我打开html文件时，它看起来像这样（但在行和列周围有边框）。非常干净，没有错误

        Ticker                         Owner    Relationship    Transaction     Cost    #Shares     Value ($)   #Shares Total   SEC Form 4
Date                                    
Sep 30  PIH     Fundamental Global Investors,       10% Owner          Buy       6.25        700         4375         352202    Sep 30 06:28 PM
Sep 28  PIH     Fundamental Global Investors,       10% Owner          Buy       6.05       36400       220220        351502    Sep 30 06:28 PM
Sep 30  FSTR                     Vizi Bradley        Director          Buy      12.00       14419       173028        801209    Sep 30 05:21 PM
Sep 29  FSTR                     Vizi Bradley        Director          Buy      12.00       11292       135504        786790    Sep 30 05:21 PM
Sep 28  FSTR                   Vizi Bradley          Director          Buy      11.83        9500       112385        775498    Sep 30 05:21 PM

现在，当我尝试通过如下代码将html文件读回数据帧时：

import pandas as pd

df =pd.read_html('Insider Trading/2016-09-30_buys.html')[0]

  Unnamed: 0 Ticker                          Owner Relationship  Transaction  \
0     Sep 30    PIH  Fundamental Global Investors,    10% Owner         Buy   
1     Sep 28    PIH  Fundamental Global Investors,    10% Owner         Buy   
2     Sep 30   FSTR                   Vizi Bradley     Director         Buy   
3     Sep 29   FSTR                   Vizi Bradley     Director         Buy   
4     Sep 28   FSTR                   Vizi Bradley     Director         Buy   

    Cost  #Shares  Value ($)  #Shares Total       SEC Form 4  Date  \
0   6.25      700       4375         352202  Sep 30 06:28 PM   NaN   
1   6.05    36400     220220         351502  Sep 30 06:28 PM   NaN   
2  12.00    14419     173028         801209  Sep 30 05:21 PM   NaN   
3  12.00    11292     135504         786790  Sep 30 05:21 PM   NaN   
4  11.83     9500     112385         775498  Sep 30 05:21 PM   NaN   

   Unnamed: 11  Unnamed: 12  Unnamed: 13  Unnamed: 14  Unnamed: 15  \
0          NaN          NaN          NaN          NaN          NaN   
1          NaN          NaN          NaN          NaN          NaN   
2          NaN          NaN          NaN          NaN          NaN   
3          NaN          NaN          NaN          NaN          NaN   
4          NaN          NaN          NaN          NaN          NaN   

   Unnamed: 16  Unnamed: 17  Unnamed: 18  Unnamed: 19  
0          NaN          NaN          NaN          NaN  
1          NaN          NaN          NaN          NaN  
2          NaN          NaN          NaN          NaN  
3          NaN          NaN          NaN          NaN  
4          NaN          NaN          NaN          NaN  

（读取html时只有一个数据帧，因此我使用[0]）

我得到的列数是原来的两倍，20列而不是10列，另外10列有“Unnamed 1”类型的名称

所以我的输出是这样的：

import pandas as pd

df =pd.read_html('Insider Trading/2016-09-30_buys.html')[0]

  Unnamed: 0 Ticker                          Owner Relationship  Transaction  \
0     Sep 30    PIH  Fundamental Global Investors,    10% Owner         Buy   
1     Sep 28    PIH  Fundamental Global Investors,    10% Owner         Buy   
2     Sep 30   FSTR                   Vizi Bradley     Director         Buy   
3     Sep 29   FSTR                   Vizi Bradley     Director         Buy   
4     Sep 28   FSTR                   Vizi Bradley     Director         Buy   

    Cost  #Shares  Value ($)  #Shares Total       SEC Form 4  Date  \
0   6.25      700       4375         352202  Sep 30 06:28 PM   NaN   
1   6.05    36400     220220         351502  Sep 30 06:28 PM   NaN   
2  12.00    14419     173028         801209  Sep 30 05:21 PM   NaN   
3  12.00    11292     135504         786790  Sep 30 05:21 PM   NaN   
4  11.83     9500     112385         775498  Sep 30 05:21 PM   NaN   

   Unnamed: 11  Unnamed: 12  Unnamed: 13  Unnamed: 14  Unnamed: 15  \
0          NaN          NaN          NaN          NaN          NaN   
1          NaN          NaN          NaN          NaN          NaN   
2          NaN          NaN          NaN          NaN          NaN   
3          NaN          NaN          NaN          NaN          NaN   
4          NaN          NaN          NaN          NaN          NaN   

   Unnamed: 16  Unnamed: 17  Unnamed: 18  Unnamed: 19  
0          NaN          NaN          NaN          NaN  
1          NaN          NaN          NaN          NaN  
2          NaN          NaN          NaN          NaN  
3          NaN          NaN          NaN          NaN  
4          NaN          NaN          NaN          NaN  

我可能做错了什么

我还根据一个建议尝试了这段代码：

df = pd.read_html('http://finviz.com/insidertrading.ashx?tc=1', header = 0, attrs = {'class': 'body-table'})[0].set_index('SEC Form 4')

---

但似乎遇到了同样的问题

如果您查看该页面的源代码，您会发现它有嵌套在表中的表，嵌套在表中。如果不做更多的挖掘，我会怀疑这是由于。您可以简单地排除所有值都为空的列。感谢您的澄清。我认为我犯了一些愚蠢的编码错误。通过指定一个属性，您也可以在开始时让数据更“干净”。例如，

attrs={'class'：'body table'}

，这似乎就是您正在尝试读取的表。谢谢，我现在就尝试一下。。奇怪的是，当我第一次将html页面读到数据框时，数据框是干净的。但是，在将其转换为html文件，然后尝试将html文件读回数据帧之后，我遇到了这种混乱。实际上，即使在添加attrs={'class'：'body table}时，我也遇到了同样的问题。我将更新这个问题。如果你查看该页面的源代码，你会发现它有嵌套在表中的表，嵌套在表中的表。如果不做更多的挖掘，我会怀疑这是由于。您可以简单地排除所有值都为空的列。感谢您的澄清。我认为我犯了一些愚蠢的编码错误。通过指定一个属性，您也可以在开始时让数据更“干净”。例如，

attrs={'class'：'body table'}

，这似乎就是您正在尝试读取的表。谢谢，我现在就尝试一下。。奇怪的是，当我第一次将html页面读到数据框时，数据框是干净的。但是，在将其转换为html文件，然后尝试将html文件读回数据帧之后，我遇到了这种混乱。实际上，即使在添加attrs={'class'：'body table}时，我也遇到了同样的问题。我会更新这个问题。