如何在带有熊猫的乳胶中输出多索引DataFrame？

Question

I am trying to output a multi-index DataFrame in a latex output using python and pandas. So far, I have this:

import pandas as pd

l = []
for a in ['1', '2', '3']:
    for b in ['first', 'second']:
        for c in ['one', 'two']:
            x = 3.2
            s = pd.Series({'a':a, 'b':b, 'c':c, 'x':x})
            l.append(pd.DataFrame(s).transpose())

df = pd.concat(l, ignore_index=True)
df = df.set_index(['a', 'b', 'c'])
print(df)
print(df.to_latex())

However, I do not have the expected output.

                x
a b      c       
1 first  one  3.2
         two  3.2
  second one  3.2
         two  3.2
2 first  one  3.2
         two  3.2
  second one  3.2
         two  3.2
3 first  one  3.2
         two  3.2
  second one  3.2
         two  3.2
\begin{tabular}{llll}
\toprule

  &       &     &    x \\
\midrule
1 & first & one &      \\
2 & second & two &  3.2 \\
\bottomrule
\end{tabular}

Is this a bug or is there something I missed?

Answer 1

I think this may be a bug in to_latex . See this question .

Based on @PaulH's comment, you could do the following:

df.reset_index().to_latex(index=False)

That will output repeated row labels, so it's more cluttered than would be ideal, but at least it outputs the whole table in LaTeX.