Data analysis is the process of filtering, manipulating, and processing raw data and datasets to get insights from them.

Python and R are usually the go-to languages for data analysis. But these days Go is becoming more and more popular for this purpose.

In this tutorial, we will be going over Gota, a data analysis package in Go, and its core functions and uses.


  • Knowledge of functional programming in Golang.
  • A Golang IDE with Go installed (I use Goland and Go 1.17.6, but you can use any other)

What is Gota?

Gota is a dataframe and data wrangling module for the Go programming language.

Gota is similar to the Pandas library in Python and is built to interface with Gonum, a scientific computing package in Golang, just like Pandas and Numpy.

The Gota module makes data wrangling (transforming and manipulating) operations in Go very easy. It works with Go inbuilt data types and various file formats, like JSON, CSV, and HTML.

Here's what we'll cover:

  • Gota Series.
  • Gota Dataframes.
  • Reading Files as Dataframes.
  • Operations On Gota Dataframes.
  • Exporting and Saving Files.

How to Get Started with Gota

Installing Gota is easy. Paste the command below into your terminal:

go get -u

This should output a successful install message. If it doesn't, update your Golang to a newer version and retry installing.

Basic Gota Concepts


Let's learn a bit about some Gota basics before diving in:

A Dataset is a collection of data, tabular or otherwise.

Dataframes are data structures that organise data into two-dimensional (rows and columns) tables, usually for the purpose of analysis.

A series is a collection of one-dimensional data belonging to a dataframe.

Note that DataFrame is the variable name of the dataframe object used as an example throughout this article.

What are Gota Series?

import ""

Gota series are created using the series.New method on compound data types, like slices and maps.

For slices, [series.New](http://series.New) takes in three arguments: it takes in the slice, the series type (type of elements to be contained in the series), and a column name.

series.New([]string{"z", "y", "d", "e"}, series.String, "col")

Series can also be made from maps by initializing the keys to be of type series and using the Type method to insert series types.

a := map[string]series.Type{
		"A": series.String,
		"D": series.Bool,

These slices can be passed into dataframes for further manipulation and operations.

What are Gota Dataframes?

Dataframe functions are contained in Gota in the dataframe sub-module.

import ""

Dataframes are data structures of other data structures. Essentially, they format the data into two-dimensional tables so you can manipulate those data. So to use dataframes, we read other data structures and data types.

We will be reading series, structs, JSON, and CSV files in this tutorial.

How to Convert a Series to a Dataframe Object

You can convert a series or set of series to a dataframe object using the [dataframe.New](http://dataframe.New) method. It takes in the series as arguments.

dataFrame := dataframe.New(
		series.New([]string{"a", "b", "c", "d", "e"}, series.String, "alphas"),
		series.New([]int{5, 4, 2, 3, 1}, series.Int, "numbers"),
		series.New([]string{"a1", "b2", "c3", "d4", "e5"}, series.String, "alnums"),
		series.New([]bool{true, false, true, true, false}, series.Bool, "state"),



alphas   numbers alnums   state
 0: a        5       a1       true
 1: b        4       b2       false
 2: c        2       c3       true
 3: d        3       d4       true
 4: e        1       e5       false
    <string> <int>   <string> <bool>

Dataframe of Structs Types

You can use structs to create dataframes.

type Dog struct {
	Name     string
	Colour      string
	Height  int
  Vaccinated  bool

dogs := []Dog{
	{"buster", "black", 56, false},
	{"jake", "white", 61, false},
	{"bingo", "brown", 50, true},
	{"gray", "cream", 68, false},

dogsDf := dataframe.LoadStructs(dogs) 

You do this by creating a slice of instances of the struct type and creating dataframes using the dataframe.Loadstructs method which takes in the slice.


Name     Colour   Height Vaccinated
 0: buster   black    56     false
 1: jake     white    61     false
 2: bingo    brown    50     true
 3: gray     cream    68     false
    <string> <string> <int>  <bool>

How to Query the Dataframe in Gota

When we have a dataframe object, we can query it for information about the composition of the dataframe using various methods.

  • dataFrame.Dims() → Outputs the dimensions of the dataframe object.
  • dataFrame.Types() → Outputs the datatypes that constitute the dataframe.
  • dataFrame.Names() → Outputs the column names of the dataframe.
  • dataFrame.Nrow() → Outputs the number of rows.
  • dataFrame.Ncol() → Outputs the number of columns.

How to Query Columns

There are many methods that come with a Gota dataframe column that help with querying column values.

  • .IsNaN() → Checks if it's a null column.
  • .Mean() → Returns the mean (average) value of the column.
  • .Copy() → Creates a new copy of the column.
  • .HasNaN() → Checks if there’s a null value in the column.
  • .Records() → returns the values in the column.
aCol := dataFrame.Col("column_name") //selects a column

How to Read Files into Dataframe Objects

JSON and CSV strings can be passed to dataframe.ReadJSON and dataframe.ReadCSV respectively.

How to Read JSON Strings

The JSON string variable is passed as an argument into dataframe.ReadJSON using strings.NewReader which returns a buffered JSON string.

	jsonString := `[
    "Name": "John",
    "Age": 44,
    "Colour": "Red",
    "Height(ft)": 6.7
    "Name": "Mary",
    "Age": 40,
    "Colour": "Blue",
    "Height(ft)": 5.7

	dataRead := dataframe.ReadJSON(strings.NewReader(jsonString))

How to Read CSV Strings

Here we have the same string in CSV format:

import (

	csvString := `
	Name, Age, Colour, Height(ft)

	dataRead := dataframe.ReadCSV(strings.NewReader(csvString))


Name      Age   Colour   Height(ft)
0: John   44    Red      6.700000
1: Mary   40    Blue     5.700000

How to Read CSV Files

Here's the CSV:


You can read CSV files by reading a file using [os.Open](http://os.Open) which takes in the file name. defer file.Close() is a context manager that helps us close the file once the program is run to prevent data loss.

	file, err := os.Open("example.csv")
	defer file.Close()
	if err != nil {
	dataFrame := dataframe.ReadCSV(file)


How to Read JSON Files

Reading JSON files is done exactly like reading CSV files.

Here's the JSON:

    "Name": "John",
    "Age": 44,
    "Colour": "Red",
    "Height(ft)": 6.7
    "Name": "Mary",
    "Age": 40,
    "Colour": "Blue",
    "Height(ft)": 5.7
    "Name": "Esther",
    "Age": 35,
    "Colour": "Black",
    "Height(ft)": 4.9
    "Name": "Mary",
    "Age": 40,
    "Colour": "Green",
    "Height(ft)": 5.2

And here's how you read the file:

	file, err := os.Open("example.json")
	defer file.Close()
	if err != nil {
	dataFrame := dataframe.ReadJSON(file)



Age   Colour   Height(ft) Name
 0: 44    Red      6.700000   John
 1: 40    Blue     5.700000   Mary
 2: 35    Black    4.900000   Esther
 3: 40    Green    5.200000   Mary
    <int> <string> <float>    <string>

Gota Dataframe Operations

How to Select Rows in Gota

You can select rows using the Subset method of the dataframe object. dataFrame.Subset takes in a slice of two integers that depict the number of rows that can be selected.

Gota provides multiple functionalities for dataframe manipulation. Using the example dataframe above, let’s go over some of these operations:

row := dataFrame.Subset([]int{0, 2})

This selects the first two rows of the dataframe.


Age   Colour   Height(ft) Name
 0: 44    Red      6.700000   John
 1: 35    Black    4.900000   Esther
    <int> <string> <float>    <string>

How to Select Columns in Gota

The Select method helps us select columns of a dataframe. [df.Select](http://df.Select) takes in a slice of two integers that depict how many columns can be selected.

column := dataFrame.Select([]int{0, 2})

We can also select columns by Index (column names) by passing a slice of strings.

column := dataFrame.Select([]string{"Name", "Colour"})


Name     Colour
 0: John     Red
 1: Mary     Blue
 2: Esther   Black
 3: Mary     Green
    <string> <string>

How to Update the Dataframe in Gota

We use the.Set method of the dataframe object to update entries. dataFrame.Set takes in a slice of integers specifying the limit of rows to be updated, and a dataframe.LoadRecords function which takes in a two-dimensional slice of the type to be passed in.

	dataFrame2 := dataFrame.Set(
		    []int{0, 3},
				[]string{"Jenny", "23", "purple", "2.2"},
				[]string{"Jesse", "34", "indigo", "3.5"},
				[]string{"Peter", "33", "violet", "3.3"},


Name     Age   Colour   Height(ft)
 0: Jesse    34    indigo   3.500000
 1: Mary     40    Blue     5.700000
 2: Esther   35    Black    4.900000
 3: Peter    33    violet   3.300000
    <string> <int> <string> <float>

How to Filter Values in Gota

To filter values, we use .Filter on the dataframe object. This takes in dataframe.F which we pass a struct literal to.

The struct literal takes in a column name Colname, a comparator Comparator, and a value Comparando which is the value for which we want to filter the dataframe.


  • series.Eq → Equal to =.
  • series.Neq → Not Equal to ≠.
  • series.Greater → Greater than >.
  • series.GreaterEq → Greater than or Equal to ≥.
  • series.Less → Less than <.
  • series.LessEq → Less than or Equal to ≤.
  • [series.In](http://series.In) → Is contained In.

In this example, we use the dataframe object from the series to the dataframe section above.

fil := dataFrame.Filter(
		dataframe.F{Colname: "alphas", Comparator: series.Eq, Comparando: "b"},


alphas   numbers alnums   state
 0: b        2       b2       false
  <string> <int>   <string> <bool>

How to Sort a Dataframe in Gota

You can sort a dataframe using the .Arrange method of the dataframe object. It takes in dataframe.Sort or dataframe.RevSort which sorst in ascending or descending order respectively. It also takes in the name of the column to be sorted as a string.

Sort by Ascending Order:

sorted := dataFrame.Arrange(


alphas   numbers alnums   state
 0: e        1       e5       false
 1: c        2       c3       true
 2: d        3       d4       true
 3: b        4       b2       false
 4: a        5       a1       true
    <string> <int>   <string> <bool>

Sort by Descending Order:

sorted := dataFrame.Arrange(

How to Use Groupby in Gota

You can use groupby to categorize data based on specific columns.

To groupby columns using Gota, we use the Groupby method and pass in the column names.

	categorise := dataFrame.GroupBy("Name", "Age")

How to Join Dataframes in Gota

Joins are a combination of dataframes. Joining dataframes with Gota works just like in SQL (Structured Query Language).

Types Of Joins:

  • Inner Join → dataFrame.InnerJoin returns a dataframe of matching values in both tables.
  • Left Join → dataFrame.LeftJoin matches similarities in the right dataframe to the left dataframe.
  • Right Join → dataFrame.RightJoin matches similarities in the left dataframe to the right dataframe.
  • Outer Join → dataFrame.OuterJoin returns all values of the dataframe.

The syntax for joining dataframes objects is:

joinVariableName := dataFrameObject.joinType(OtherDataframe, JoinKey)

The Join key is the column of the dataframe object where the join is to be executed.

Example of a Left Join

 df := dataframe.New(
		series.New([]string{"a", "b", "c", "d", "e"}, series.String, "alphas"),
		series.New([]int{5, 4, 2, 3, 1}, series.Int, "numbers"),
		series.New([]string{"a1", "b2", "c3", "d4", "e5"}, series.String, "alnums"),
		series.New([]bool{true, false, true, true, false}, series.Bool, "state"),
	df2 := dataframe.New(
		series.New([]string{"f", "g", "h", "i", "j"}, series.String, "alphas"),
		series.New([]int{1, 2, 3, 4, 5}, series.Int, "numbers"),
		series.New([]string{"f6", "g7", "h8", "i9", "j10"}, series.String, "alnums"),
		series.New([]bool{false, true, false, false, true}, series.Bool, "state"),
	join := df.RightJoin(df2, "state")


[12x7] DataFrame

    state  alphas_0 numbers_0 alnums_0 alphas_1 numbers_1 alnums_1
 0: false  b        4         b2       f        1         f6
 1: false  e        1         e5       f        1         f6
 2: true   a        5         a1       g        2         g7
 3: true   c        2         c3       g        2         g7
 4: true   d        3         d4       g        2         g7
 5: false  b        4         b2       h        3         h8
 6: false  e        1         e5       h        3         h8
 7: false  b        4         b2       i        4         i9
 8: false  e        1         e5       i        4         i9
 9: true   a        5         a1       j        5         j10
    ...    ...      ...       ...      ...      ...       ...
    <bool> <string> <int>     <string> <string> <int>     <string>

How to Apply Functions to a Dataframe in Gota

To apply functions to columns and rows of a dataframe, we use Capply and Rapply, respectively. These take in the function to be applied on the column or row.


How to Use Describe on a Dataframe in Gota

Using Describe() on a dataframe object returns descriptive statistics on the values of the dataframe.

description := dataFrame.Describe()


column   alphas   numbers  alnums   state
 0: mean     -        3.000000 -        0.600000
 1: median   -        3.000000 -        NaN
 2: stddev   -        1.581139 -        0.547723
 3: min      a        1.000000 a1       0.000000
 4: 25%      -        2.000000 -        0.000000
 5: 50%      -        3.000000 -        1.000000
 6: 75%      -        4.000000 -        1.000000
 7: max      e        5.000000 e5       1.000000
    <string> <string> <float>  <string> <float>

How to Export Dataframes (Writing Files in Go)

We export the manipulated data using the WriteCSV method of the dataframe object. The dataFrame.WriteCSV takes in a file name which it creates or inserts into.

file, err := os.Create("output.csv")
	if err != nil {


To export JSON, we use dataFrame.WriteJSON in the same way.


In this tutorial, you have learned how to perform data analysis in Go. You've also learned about the Gota package's various functions.

It's still a good idea to primarily use Python and R for data analysis as they are considered the industry standard. But Gota is useful for applications requiring speed and homogeneity.

Check out the Gota documentation to learn more or contribute to the project.

Have fun coding and exploring!