Notebook display

What a frame prints, and the HTML table a notebook shows, as go-dataframe v0.1.0 produced them.

DataFrame, Series and Index have Display(), which returns the value in every format a notebook front end may show, keyed by MIME type: text/plain (what String prints, laid out as pandas prints it) and text/html (the table pandas shows in a notebook). It is the display contract of the go-notebook kernel, so a cell whose last expression is a frame shows it as a table.

Every output on this page was produced by go-dataframe v0.1.0 from the code shown above it: figures/<name>.go in this site’s repository writes it, and the CI writes it again and fails if one byte differs.

A frame

The table the examples share, built from Go slices:

package main

func init() { registerOutput("display", display) }

// What a notebook cell whose value is a frame shows: Display returns both
// MIME types, text/plain (String) and text/html (HTML).
func display() (string, string) {
	df := rainfall()
	d := df.Display()
	return string(d["text/plain"]), string(d["text/html"])
}

Printed (text/plain):

    city  year   rain  days
0   Lyon  2022  741.2    98
1  Paris  2022  512.9   103
2   Nice  2022  698.4    61
3   Lyon  2023  830.5   112
4  Paris  2023  640.2   118
5   Nice  2023  702.3    64
6   Lyon  2024  810.0   107
7  Paris  2024  598.7   111
8   Nice  2024  760.1    70

Shown in a notebook (text/html):

	city	year	rain	days
0	Lyon	2022	741.2	98
1	Paris	2022	512.9	103
2	Nice	2022	698.4	61
3	Lyon	2023	830.5	112
4	Paris	2023	640.2	118
5	Nice	2023	702.3	64
6	Lyon	2024	810.0	107
7	Paris	2024	598.7	111
8	Nice	2024	760.1	70

A group-by

package main

import "github.com/go-dataframe/dataframe"

func init() { registerOutput("groupby", groupby) }

// Group by city: the mean and the maximum yearly rainfall, and the rainy days
// summed over the three years.
func groupby() (string, string) {
	g := dataframe.Must(rainfall().GroupBy("city"))
	out := dataframe.Must(g.Agg(
		dataframe.Agg("rain", "mean", "max"),
		dataframe.Agg("days", "sum"),
	))
	return out.String(), out.HTML()
}

Printed (text/plain):

        rain_mean  rain_max  days
city                             
Lyon   793.900000     830.5   317
Nice   720.266667     760.1   195
Paris  583.933333     640.2   332

Shown in a notebook (text/html):

	rain_mean	rain_max	days
city
Lyon	793.900000	830.5	317
Nice	720.266667	760.1	195
Paris	583.933333	640.2	332

Describe

package main

func init() { registerOutput("describe", describe) }

// df.describe(): count, mean, std, min, quartiles and max of each numeric column.
func describe() (string, string) {
	d := rainfall().Describe()
	return d.String(), d.HTML()
}

Printed (text/plain):

              year        rain        days
count     9.000000    9.000000    9.000000
mean   2023.000000  699.366667   93.777778
std       0.866025  102.034357   22.415273
min    2022.000000  512.900000   61.000000
25%    2022.000000  640.200000   70.000000
50%    2023.000000  702.300000  103.000000
75%    2024.000000  760.100000  111.000000
max    2024.000000  830.500000  118.000000

Shown in a notebook (text/html):

	year	rain	days
count	9.000000	9.000000	9.000000
mean	2023.000000	699.366667	93.777778
std	0.866025	102.034357	22.415273
min	2022.000000	512.900000	61.000000
25%	2022.000000	640.200000	70.000000
50%	2023.000000	702.300000	103.000000
75%	2024.000000	760.100000	111.000000
max	2024.000000	830.500000	118.000000

How the text layout and the HTML are produced, and where they follow pandas, is in Design.

Created on October 11, 2026 by tannevaled

Modified on October 11, 2026 by tannevaled

The API, call by call Plots