How to Convert a Julia DataFrame to a LaTeX Table Similar to the pandas.to_latex()?

I am trying to convert a DataFrame in Julia to a LaTeX table, similar to how it is done in Python.

I have examples of both the Python and Julia code. In Python, I use pandas to convert the DataFrame to a LaTeX table with specific formatting options. Here's the Python code:

import pandas as pd

df = pd.DataFrame(
    dict(name=["Raphael", "Donatello"], age=[26, 45], height=[181.23, 177.65])
)

print(
    df.to_latex(
        index=False,
        formatters={"name": str.upper},
        float_format="{:.1f}".format,
    )
)

The output LaTeX table from the above Python code looks like this:

\begin{tabular}{lrr}
\toprule
name & age & height \\
\midrule
Raphael & 26 & 181.23 \\
Donatello & 45 & 177.65 \\
\bottomrule
\end{tabular}

https://i.sstatic.net/KnBA0iBG.png

In Julia, I have found the following code:

using DataFrames

df = DataFrame(name=["Raphael", "Donatello"], age=[26, 45], height=[181.23, 177.65])

show(stdout, MIME("text/latex"), df)

This code generates a basic LaTeX table that looks like this:

\begin{tabular}{r|ccc}
    & name & age & height\\
    \hline
    & String & Int64 & Float64\\
    \hline
    1 & Raphael & 26 & 181.23 \\
    2 & Donatello & 45 & 177.65 \\
\end{tabular}

https://i.sstatic.net/J7pnoV2C.png

But, I would like to know how to apply similar formatting as in the Python example.

Solution

There are a couple of options. For example:

    using Pkg
    
    pkg"activate --temp"
    pkg"add DataFrames, Latexify"
    
    using DataFrames, Latexify
    
    df = DataFrame(name = ["Raphael", "Donatello"], age = [26, 45], height = [181.23, 177.65])
    
    # Don't forget to read the docs, including
    # https://korsbo.github.io/Latexify.jl/stable/arguments/#Tabular
    latexify(df; env = :table, booktabs = true, latex = false) |> print

which will give you

    \begin{tabular}{ccc}
    \toprule
    name & age & height\\
    \midrule
    Raphael & 26 & 181.23\\
    Donatello & 45 & 177.65\\
    \bottomrule
    \end{tabular}

you can further manipulate the resulting string (e.g. to change the position specifier to lrr)

    l = latexify(df; env = :table, booktabs = true, latex = false)
    using LaTeXStrings
    latexstring(replace(l, "{ccc}" => "{lrr}"))

(some functions operating on Strings are also defined to work with LaTeXStrings - see this)

Or, if you want more control over the output to match the Python version:

    pkg"add LaTeXTabulars, LaTeXStrings"
    
    using LaTeXTabulars
    using LaTeXStrings

    # https://github.com/tpapp/LaTeXTabulars.jl
    latex_tabular(
        "mytable.tex",
        Tabular("lrr"),
        [Rule(:top), names(df), Rule(:mid), df[1, :], df[2, :], Rule(:bottom)],
    )

— which you can also do programmatically — to get:

    \begin{tabular}{lrr}
    \toprule 
    name & age & height \\
    \midrule 
    Raphael & 26 & 181.23 \\
    Donatello & 45 & 177.65 \\
    \bottomrule 
    \end{tabular}

EDIT: you can further configure the latexify call, just like you did df.to_latex. You can specify the formatter, add it as default, etc.