Python Pandas read_csv無法正確導入

Question

我有一個.xls文件，看起來與此類似...

Value of Construction Put in Place...
(Millions of Dollars....)
Blank Row
Date    Total_Construction Total Residential Total Nonresidential...Columns 
Dec-15  1,116,570          435,454           681,217 
Nov-15  1,115,966          432,295           683,671
Oct-15  1,122,749          431,164           691,585   
.
.
.

我正在嘗試導入文件以獲取以下信息：

Date    Total_Construction Total Residential Total Nonresidential 
Dec-15  1,116,570          435,454           681,217 
Nov-15  1,115,966          432,295           683,671
Oct-15  1,122,749          431,164           691,585   
.
.
. 

使用以下代碼：

for chunk in pandas.read_csv('/PATH/totsatime.xls',
                 names      = ['Date', 'Total Residential', 'Total Nonresidential'],
                 header     = 4,
                 chunksize  = 1,
                 skiprows   = range(1, 4),
                 thousands  = ','):

    if chunk['Date'] == 'Dec-01':
        break

    else:
        df = pandas.DataFrame(chunk)

但是，我最終得到以下結果：

Date             Total Residential     Total Nonresidential
Lodging          NaN                   NaN
Office          NaN                   NaN
Commercial      NaN                   NaN
Health care     NaN                   NaN

日期最終從我不導入的列中格式化。 任何建議將不勝感激。

先感謝您。

Answer 1

不要使用read_csv導入xls文件。 使用read_excel 。 參見http://pandas.pydata.org/pandas-docs/stable/generated/pandas.read_excel.html