在 function 中将 pandas 列的值作为单独的 arguments 传递

Question

I have a csv file that has rows xy and z and columns of different names. Essentially these are three-dimensional coordinates of each name. I have imported this csv as a dataframe which looks like:

 Coordinate       C1       C2       C3       C4       C5       N6
0          x  0.16620  0.20640  0.16240  0.08140  0.04370  0.08288
1          y  0.22340  0.34680  0.44090  0.41100  0.28550  0.18996
2          z  0.38187  0.42618  0.40091  0.33013  0.28793  0.31430

I would like to perform transformations to the coordinates such as (x,y,z)->(y,x,z). I don't have much experience with python but I find I can do this with a simple function such as

def trans(x,y,z):
     return (y,x,z)

I am having issues with getting the values from each column in the form of x, y, and z so that I may pass them through the function to achieve the desired transformation. I find that I can convert each column to a tuple by using

tuple([tuple(co_df[col]) for col in co_df])

however, this will ultimately give me a tuple of tuples. I could instead turn this into a list of tuples, but then I have the issue of passing a list of tuples into my function to transform the coordinates. Any help is appreciated!

Answer 1

I don't have enough reputation to comment yet, but maybe a for loop would be useful

For example, to print the output of trans for each column, you could do something like this:

for column in list(co_df):
    a,b,c=co_df[column]
    print(trans(a,b,c))

This would print:

('y', 'x', 'z')
(0.22340, 0.16620, 0.38187)
(0.34680, 0.20640, 0.42618)
(0.44090, 0.16240, 0.40091)
(0.41100, 0.08140, 0.33013)
(0.28550, 0.04370, 0.28793)
(0.18996, 0.08288, 0.31430)

What are you planning on doing with the values once they been transformed? Do you need them stored in an object?

EDIT Re. your question about what a,b,c=co_df[column] does, think about this at a single column level eg co_df["C1"] print(co_df["C1"]) returns

0    0.16620
1    0.22340
2    0.38187

Doing a,b,c=co_df["C1"] assigns each value of co_df["C1"] to a, b, and c respectively.

Answer 2

Here is the original data frame:

from io import StringIO
import pandas as pd
data = ''' Coordinate       C1       C2       C3       C4       C5       N6
0          x  0.16620  0.20640  0.16240  0.08140  0.04370  0.08288
1          y  0.22340  0.34680  0.44090  0.41100  0.28550  0.18996
2          z  0.38187  0.42618  0.40091  0.33013  0.28793  0.31430
'''

df = pd.read_csv(StringIO(data), sep='\s+')
print(df)

  Coordinate       C1       C2       C3       C4       C5       N6
0          x  0.16620  0.20640  0.16240  0.08140  0.04370  0.08288
1          y  0.22340  0.34680  0.44090  0.41100  0.28550  0.18996
2          z  0.38187  0.42618  0.40091  0.33013  0.28793  0.31430

First, you can put the columns in a different order like this:

print(df[['Coordinate', 'N6', 'C5', 'C4', 'C3', 'C2', 'C1']])

  Coordinate       N6       C5       C4       C3       C2       C1
0          x  0.08288  0.04370  0.08140  0.16240  0.20640  0.16620
1          y  0.18996  0.28550  0.41100  0.44090  0.34680  0.22340
2          z  0.31430  0.28793  0.33013  0.40091  0.42618  0.38187

Second, you can re-label the columns like this (assign to df.columns):

df.columns = ['Coordinate', 'N6', 'C5', 'C4', 'C3', 'C2', 'C1']
print(df)

  Coordinate       N6       C5       C4       C3       C2       C1
0          x  0.16620  0.20640  0.16240  0.08140  0.04370  0.08288
1          y  0.22340  0.34680  0.44090  0.41100  0.28550  0.18996
2          z  0.38187  0.42618  0.40091  0.33013  0.28793  0.31430