Pandas: How could I iterate two dataframes which have exactly same format?

问题

My final goal is making list which contain a pair for corresponding location of dataframes, like below

 [df_one_first_element, df_two_first_element, column_first, index_first]

 :[0.619159, 0.510162, 20140109,0.50], [0.264191,0.269053,20140213,0.50]...

So I am trying to iterate two dataframe but got stuck now. How could I iterate two dataframe which has exactly same format but different data.

For example, I have two dataframes; df_one and df_two that appear like the below:

df_one = 

      20140109  20140213  20140313  20140410  20140508  20140612  20140710  \
0.50  0.619159  0.264191  0.438849  0.465287  0.445819  0.412582  0.397366   
0.55  0.601379  0.303953  0.457524  0.432335  0.415333  0.382093  0.382361  

df_two = 

      20140109  20140213  20140313  20140410  20140508  20140612  20140710  \
0.50  0.510162  0.269053  0.308494  0.300554  0.294360  0.286980  0.280494   
0.55  0.489953  0.258690  0.290044  0.283933  0.278180  0.271426  0.266580

And I want to access the same location of the dataframe by iterating over the whole values in the dataframe.

Firstly I tried iterrows()

i = 0
for index, row in df_one.iterrows():
    j= 0
    for item in row:
        print df_two(i,j)
        j= j+1
    i = i+1

but as you know we can not access like:

df_two(i,j)

So I am currently lost the way. Or could we access the data by index name and column name?

回答1:

Below code will also enable you to find values on both dataframes in same locations.

python 2x

for i in range(0, len(df_one.index)):
    for j in range(0, len(df_one.columns)):
        print df_one.values[i,j],df_two.values[i,j],i,j

python 3x

for i in range(0, len(df_one.index)):
    for j in range(0, len(df_one.columns)):
        print(df_one.values[i,j],df_two.values[i,j],i,j)

回答2:

You could use itertools.izip :

for ( idxRow, s1 ), ( _, s2 ) in itertools.izip( df0.iterrows(), df1.iterrows() ) :
    for ( idxCol, v1 ), ( _, v2 ) in itertools.izip( s1.iteritems(), s2.iteritems() ) :
        print ( v1, v2, idxCol, idxRow )

In:

X   Y   Z
a    1.171124    0.853229    1.416635
b    0.971665   -1.727410   -0.055180

Out:

(1.1711241491561419, 1.3715317727366974, 'X', 'a')
(0.85322862359611618, 0.72799908412372294, 'Y', 'a')
(1.4166350896829785, 2.0068549773211006, 'Z', 'a')
(0.9716653056530119, 0.94413346620976102, 'X', 'b')
(-1.727409829928936, 2.9839447205351157, 'Y', 'b')
(-0.055180403519242693, 0.0030448769325464513, 'Z', 'b')

回答3:

I solved this problem by get_value mehtod

http://pandas.pydata.org/pandas-docs/version/0.8.1/indexing.html

Here is my code it looks working

df_columns = df_one.columns.values
for index, row in df_one.iterrows():
    j= 0
    for item in row:
        print df_two.get_value(index, df_columns[j])
        j= j+1

来源：https://stackoverflow.com/questions/24709557/pandas-how-could-i-iterate-two-dataframes-which-have-exactly-same-format

标签

python

pandas