0
我想有一個腳本通過行去行中使用Python（熊貓）。我希望它如果列A ='東西'和列B> 25，然後寫入列C =「類別」。逐行（熊貓） - 如果列A =「東西」和列B> 25然後列C =「類別」
我有一個月列和列日。因此，舉例來說：
當月=月和日> = 25則周= 8月25日
我試了幾件事情，既不工作...
首先我想：
import os ### OS library is imported.
import pandas as pd ### Pandas library is imporated as 'pd'.
counter = 1 ### Counter starts at the first iteration.
while os.path.exists("CSV-Iteration-"'{0}'"/".format(counter)): ### Runs the loop until all iteration's folders have been processed.
a = pd.read_csv("output-"'{0}'".csv".format(counter)) ### Sets 'a' dataframe as holding data from a CSV file.
a['Week'] = ""
a[(a['Month'] is 'June') & (a['Day'] < 25)]['Week'] = 'June 18'
a[(a['Month'] is 'June') & (a['Day'] >= 25)]['Week'] = 'June 25'
a[(a['Month'] is 'July') & (a['Day'] < 2)]['Week'] = 'June 25'
a[(a['Month'] is 'July') & (a['Day'] >= 2) & (a['Day'] < 9)]['Week'] = 'July 2'
a[(a['Month'] is 'July') & (a['Day'] >= 9) & (a['Day'] < 16)]['Week'] = 'July 9'
a[(a['Month'] is 'July') & (a['Day'] >= 16) & (a['Day'] < 23)]['Week'] = 'July 16'
a[(a['Month'] is 'July') & (a['Day'] >= 23) & (a['Day'] < 30)]['Week'] = 'July 23'
a[(a['Month'] is 'July') & (a['Day'] >= 31) & (a['Day'] < 16)]['Week'] = 'July 30'
a[(a['Month'] is 'August') & (a['Day'] < 6)]['Week'] = 'July 30'
a[(a['Month'] is 'August') & (a['Day'] >= 6) & (a['Day'] < 13)]['Week'] = 'August 6'
a[(a['Month'] is 'August') & (a['Day'] >= 13) & (a['Day'] < 20)]['Week'] = 'August 13'
a[(a['Month'] is 'August') & (a['Day'] >= 20) & (a['Day'] < 27)]['Week'] = 'August 20'
a[(a['Month'] is 'August') & (a['Day'] >= 27)]['Week'] = 'August 27'
a[(a['Month'] is 'September') & (a['Day'] < 3)]['Week'] = 'August 27'
a[(a['Month'] is 'September') & (a['Day'] >= 3) & (a['Day'] < 10)]['Week'] = 'September 3'
a[(a['Month'] is 'September') & (a['Day'] >= 10) & (a['Day'] < 17)]['Week'] = 'September 10'
a[(a['Month'] is 'September') & (a['Day'] >= 17) & (a['Day'] < 24)]['Week'] = 'September 17'
a[(a['Month'] is 'September') & (a['Day'] >= 24)] = 'September 24'
a[(a['Month'] is 'October') & (a['Day'] >= 1) & (a['Day'] < 8)]['Week'] = 'October 1'
a[(a['Month'] is 'October') & (a['Day'] >= 8) & (a['Day'] < 15)]['Week'] = 'October 8'
a[(a['Month'] is 'October') & (a['Day'] >= 15) & (a['Day'] < 22)]['Week'] = 'October 15'
a[(a['Month'] is 'October') & (a['Day'] >= 22) & (a['Day'] < 29)]['Week'] = 'October 22'
a[(a['Month'] is 'October') & (a['Day'] >= 29)]['Week'] = 'October 29'
a[(a['Month'] is 'November') & (a['Day'] < 5)]['Week'] = 'October 29'
a[(a['Month'] is 'November') & (a['Day'] >= 5) & (a['Day'] < 12)]['Week'] = 'November 5'
a[(a['Month'] is 'November') & (a['Day'] >= 12) & (a['Day'] < 19)]['Week'] = 'November 12'
a[(a['Month'] is 'November') & (a['Day'] >= 19) & (a['Day'] < 26)]['Week'] = 'November 19'
a[(a['Month'] is 'November') & (a['Day'] >= 26)]['Week'] = 'November 26'
a[(a['Month'] is 'December') & (a['Day'] < 3)]['Week'] = 'November 26'
a[(a['Month'] is 'December') & (a['Day'] >= 3) & (a['Day'] < 10)]['Week'] = 'December 3'
a[(a['Month'] is 'December') & (a['Day'] >= 10) & (a['Day'] < 17)]['Week'] = 'December 10'
a[(a['Month'] is 'December') & (a['Day'] >= 17) & (a['Day'] < 24)]['Week'] = 'December 17'
a[(a['Month'] is 'December') & (a['Day'] >= 24) & (a['Day'] < 31)]['Week'] = 'December 24'
a[(a['Month'] is 'December') & (a['Day'] >= 31)]['Week'] = 'December 31'
a[(a['Month'] is 'January') & (a['Day'] < 7)]['Week'] = 'December 31'
a[(a['Month'] is 'January') & (a['Day'] >= 7) & (a['Day'] < 14)]['Week'] = 'January 7'
a[(a['Month'] is 'January') & (a['Day'] >= 14) & (a['Day'] < 21)]['Week'] = 'January 14'
a[(a['Month'] is 'January') & (a['Day'] >= 21) & (a['Day'] < 28)]['Week'] = 'January 21'
a[(a['Month'] is 'January') & (a['Day'] >= 28)]['Week'] = 'January 28'
a.to_csv("TESToutput-"'{0}'".csv".format(counter), index=False) ### 'a' dataframe becomes 'TESToutput-#.csv' and does not print fields for indexing (index=False).
counter += 1 ### Adds 1 to the counter.
print 'Date Corrections - All Done!'
然後我嘗試：
import os ### OS library is imported.
import pandas as pd ### Pandas library is imporated as 'pd'.
counter = 1 ### Counter starts at the first iteration.
while os.path.exists("CSV-Iteration-"'{0}'"/".format(counter)): ### Runs the loop until all iteration's folders have been processed.
a = pd.read_csv("output-"'{0}'".csv".format(counter)) ### Sets 'a' dataframe as holding data from a CSV file.
a['Week'] = ""
def this_week (row):
if row[(a['Month'] is 'June') + (a['Day'] < 25)]:
return 'June 18'
if row[(a['Month'] is 'June') + (a['Day'] >= 25)]:
return 'June 25'
if row[(a['Month'] is 'July') + (a['Day'] < 2)]:
return 'June 25'
if row[(a['Month'] is 'July') + (a['Day'] >= 2) + (a['Day'] < 9)]:
return 'July 2'
if row[(a['Month'] is 'July') + (a['Day'] >= 9) + (a['Day'] < 16)]:
return 'July 9'
if row[(a['Month'] is 'July') + (a['Day'] >= 16) + (a['Day'] < 23)]:
return 'July 16'
if row[(a['Month'] is 'July') + (a['Day'] >= 23) + (a['Day'] < 30)]:
return 'July 23'
if row[(a['Month'] is 'July') + (a['Day'] >= 31) + (a['Day'] < 16)]:
return 'July 30'
if row[(a['Month'] is 'August') + (a['Day'] < 6)]:
return 'July 30'
if row[(a['Month'] is 'August') + (a['Day'] >= 6) + (a['Day'] < 13)]:
return 'August 6'
if row[(a['Month'] is 'August') + (a['Day'] >= 13) + (a['Day'] < 20)]:
return 'August 13'
if row[(a['Month'] is 'August') + (a['Day'] >= 20) + (a['Day'] < 27)]:
return 'August 20'
if row[(a['Month'] is 'August') + (a['Day'] >= 27)]:
return 'August 27'
if row[(a['Month'] is 'September') + (a['Day'] < 3)]:
return 'August 27'
if row[(a['Month'] is 'September') + (a['Day'] >= 3) + (a['Day'] < 10)]:
return 'September 3'
if row[(a['Month'] is 'September') + (a['Day'] >= 10) + (a['Day'] < 17)]:
return 'September 10'
if row[(a['Month'] is 'September') + (a['Day'] >= 17) + (a['Day'] < 24)]:
return 'September 17'
if row[(a['Month'] is 'September') + (a['Day'] >= 24)]:
return 'September 24'
if row[(a['Month'] is 'October') + (a['Day'] >= 1) + (a['Day'] < 8)]:
return 'October 1'
if row[(a['Month'] is 'October') + (a['Day'] >= 8) + (a['Day'] < 15)]:
return 'October 8'
if row[(a['Month'] is 'October') + (a['Day'] >= 15) + (a['Day'] < 22)]:
return 'October 15'
if row[(a['Month'] is 'October') + (a['Day'] >= 22) + (a['Day'] < 29)]:
return 'October 22'
if row[(a['Month'] is 'October') + (a['Day'] >= 29)]:
return 'October 29'
if row[(a['Month'] is 'November') + (a['Day'] < 5)]:
return 'October 29'
if row[(a['Month'] is 'November') + (a['Day'] >= 5) + (a['Day'] < 12)]:
return 'November 5'
if row[(a['Month'] is 'November') + (a['Day'] >= 12) + (a['Day'] < 19)]:
return 'November 12'
if row[(a['Month'] is 'November') + (a['Day'] >= 19) + (a['Day'] < 26)]:
return 'November 19'
if row[(a['Month'] is 'November') + (a['Day'] >= 26)]:
return 'November 26'
if row[(a['Month'] is 'December') + (a['Day'] < 3)]:
return 'November 26'
if row[(a['Month'] is 'December') + (a['Day'] >= 3) + (a['Day'] < 10)]:
return 'December 3'
if row[(a['Month'] is 'December') + (a['Day'] >= 10) + (a['Day'] < 17)]:
return 'December 10'
if row[(a['Month'] is 'December') + (a['Day'] >= 17) + (a['Day'] < 24)]:
return 'December 17'
if row[(a['Month'] is 'December') + (a['Day'] >= 24) + (a['Day'] < 31)]:
return 'December 24'
if row[(a['Month'] is 'December') + (a['Day'] >= 31)]:
return 'December 31'
if row[(a['Month'] is 'January') + (a['Day'] < 7)]:
return 'December 31'
if row[(a['Month'] is 'January') + (a['Day'] >= 7) + (a['Day'] < 14)]:
return 'January 7'
if row[(a['Month'] is 'January') + (a['Day'] >= 14) + (a['Day'] < 21)]:
return 'January 14'
if row[(a['Month'] is 'January') + (a['Day'] >= 21) + (a['Day'] < 28)]:
return 'January 21'
if row[(a['Month'] is 'January') + (a['Day'] >= 28)]:
return 'January 28'
a['Week'] = a.apply (lambda row: this_week (row), axis=1)
a.to_csv("TESToutput-"'{0}'".csv".format(counter), index=False) ### 'a' dataframe becomes 'TESToutput-#.csv' and does not print fields for indexing (index=False).
counter += 1 ### Adds 1 to the counter.
print 'Date Corrections - All Done!'
第二個給了我這個錯誤：「IndexingError：（'聯合國提供可對準的布爾系列鑰匙」，在索引0' u'occurred）」
我很新的Python的，所以我把這些結合在一起基於我在論壇上看過。請讓我知道是否有一個更簡單的方法來做到這一點，或者如果有更正或補充，使這兩個腳本之一工作。
謝謝！
============================================== ======
BREAK - 最新資訊低於
這是一個數據文件的樣子（減去多餘的列）。
Month Day C.Sym F.Sym D.Sym
September 3 1 1
September 27 1
October 14 1
October 15 1
October 17 1
October 21 1
October 29 1
November 30 1
December 16 1 1
December 17 1
December 27 1
January 6 1
January 8 1
January 20 1
我想添加一個檢查月份和日期列以分配「周」IE的列。下圖：
Month Day C.Sym F.Sym D.Sym Week
September 3 1 1 Sept 3
September 27 1 Sept 24
October 14 1 Oct 8
October 15 1 Oct 15
October 17 1 Oct 15
October 21 1 Oct 15
October 29 1 Oct 29
November 30 1 Oct 29
December 16 1 1 Dec 10
December 17 1 Dec 10
December 27 1 Dec 24
January 6 1 Dec 31
January 8 1 Jan 7
January 20 1 Jan 14
的elif的，我想現在納入的一個例子：
elif a[(a['Month'] is 'January') & (a['Day'] >= 14) & (a['Day'] < 21)]:
['Week'] = 'January 14'
我希望這是更具體的幫助...
我的問題我遇到的問題是我的答案需要超過True/False。我需要能夠把「6月18日」，「6月25日」，「7月2日」等取決於Column_Month和Column_Day值。現在，我想我需要運行一個內部的多個ifelse查詢另一個開始的最早日期和結束最晚日期實現我的目標......還是我失去了「東西」這個答案嗎？ –
@DavidMills你必須具體。如果你發佈了你的數據目前的樣子和你想要的輸出結果，你更有可能得到解決問題的答案。 – spies006
我對上面的信息添加了註釋，數據看起來像什麼以及我想要的輸出是什麼。謝謝！ –