Pandas: Using variables to create dataframe with one row and column names from variable names

Question

Suppose I have some variables in Python. I am trying to create a 1-row Pandas dataframe, where the column names are the variables' names and the values in the row are from the variables.

For example, if I have this code:

pi  = 3.142
e   = 2.718
phi = 1.618

I would like a dataframe that conceptually looks like this:

     pi     e      phi
0   3.142   2.718  1.618

I tried the following, but everything is in one column, and the variable names are not added:

df = pd.DataFrame(data=[pi, e, phi])
df
#        0
# 0  3.140
# 1  2.718
# 2  1.618

Note that I'm trying to replicate the behavior of some of my older R code.

pi  <- 3.142
e   <- 2.718
phi <- 1.618
df  <- data.frame(pi, e, phi)
df
#      pi     e   phi
# 1 3.142 2.718 1.618

Answer 1

I think you were looking for this format:

pd.DataFrame([[pi,e,phi]],columns=['pi','e','phi'])

Output:

      pi      e    phi
0  3.142  2.718  1.618

Answer 2

You can use a list and dictionary like this

df = pd.DataFrame([{'pi':pi, 'e':e, 'phi':phi}])

which returns

df
Out[5]: 
       e    phi     pi
0  2.718  1.618  3.142

to preserve the column order, you can use the columns argument:

df = pd.DataFrame([{'pi':pi, 'e':e, 'phi':phi}], columns=['pi', 'e', 'phi'])

which returns

df
Out[9]: 
      pi      e    phi
0  3.142  2.718  1.618

Additional rows would go into separate dictionaries like this:

df = pd.DataFrame([{'pi':pi, 'e':e, 'phi':phi}, {'pi':2, 'e':3, 'phi':1}])

Answer 3

#Reference columns names only once and column order is retained.    
pd.concat([pd.DataFrame(data=[eval(k)],columns=[k]) for k in ['pi','e','phi']],axis=1)
Out[1226]: 
      pi      e    phi
0  3.142  2.718  1.618