current community

your communities

more stack exchange communities

Stack Exchange
tour help
careers 2.0
Take the 2-minute tour ×
Stack Overflow is a question and answer site for professional and enthusiast programmers. It's 100% free, no registration required.
up vote 26 down vote favorite
9

I have a DataFrame with named columns and rows indexed with not continuous numbers like from the code:

df1 = DataFrame(np.random.randn(10, 4), columns=['a', 'b', 'c', 'd'])
mask = df1.applymap(lambda x: x <-0.7)
df1 = df1[-mask.any(axis=1)]
sLength = len(df1['a'])
e = Series(np.random.randn(sLength))

I would like to add new column 'e' to the existing df and do not change anything in the df. (The series got always the same length as a dataframe.) I try different version of join, append, merge but do not have this what I want, error at the most.

The series and df is already given and above code is only to illustrate example.

I am sure there is some easy way to that but can't figure it out

share|improve this question
add comment

3 Answers

up vote 38 down vote accepted

Use the original df1 indexes to create the series:

df1['e'] = Series(np.random.randn(sLength), index=df1.index)
share|improve this answer
    
The series comes from sensor and are fed to the computer. The only thing is that it has given length, the same length as DataFrame. The presented code is only to illustrate example –  tomasz74 Sep 23 '12 at 19:29
    
@tomasz74 Not sure what do you mean and how that affects your question and my answer. –  joaquin Sep 23 '12 at 19:34
    
Thanks a lot @joaquin your answer is perfectly what I couldn't figure out. –  tomasz74 Sep 23 '12 at 19:52
1  
if you need to prepend column use DataFrame.insert: df1.insert(0, 'A', Series(np.random.randn(sLength), index=df1.index)) –  lowtech Dec 9 '13 at 21:48
add comment
up vote 14 down vote

This is the simple way of adding a new column: df['e'] = e

share|improve this answer
add comment
up vote 8 down vote

One way to do it would be to use map:

df1['e'] = df1['a'].map(lambda x: np.random.random())

Edit: Saying that, doing this directly is via numpy will be much more efficient:

df1['e'] = np.random.randn(sLength)
share|improve this answer
    
thanks for your reply, as I have e already given, have can I modify your code, .map to use existing series instead of lambda? I try df1['e'] = df1['a'].map(lambda x: e) or df1['e'] = df1['a'].map(e) but it not what I need. (I am new to pyhon and your previous answer already helped me) –  tomasz74 Sep 23 '12 at 20:03
    
@tomasz74 if you already have e as a Series then you don't need to use map, use df['e']=e (@joaquins answer). –  Andy Hayden Sep 23 '12 at 20:33
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.