Distinctive Analytics

This library is named Plotly after the company of the same name. Plotly provides visualization libraries for Python, R, MATLAB, Perl, Julia, Arduino, and REST.

Installation

Install the library in the command prompt if on the system, or you can directly run and install the command in a cell if you are using Google Colab or Jupyter Lab.

!pip install plotly

install.packages('plotly')

Importing Plotly

Syntax for importing plotly.

# direct import 
import plotly 
# specific import 
import plotly.express as px

library(plotly)

Well, we understood the installation and importing parts, now let’s learn how to use the library to carry out your visualizations.

Data:

Observe the data sample above. We have two columns. First are years of experience, and second is salary. Salary is dependent on experience.

Python

R

Python

Python

R

import plotly.express as px

# provide data to line plot function in px.line
fig=px.line(x=[1,2,3,4],y=[4,6,9,12])

# plot the line
fig.show()

library(plotly)
data_line <- data.frame(x=c(1,2,3,4),y=c(4,6,9,12))
fig <- plot_ly(data_line, x = ~x, y = ~y, mode = 'lines')

Line plots with the Iris dataset in line plot px. The line represents the vertices of each data point.

Python

R

Python

Python

R

# import plotly express library
import plotly.express as px

# Load the stocks dataset
df = px.data.stocks()

# plotting the line chart
fig = px.line(df, x="date", y="GOOG")

# showing the plot
fig.show()

# import plotly express library
import plotly.express as px

# Load the stocks dataset
df = px.data.stocks()

# plotting the line chart
fig = px.line(df, x="date", y="GOOG")

# showing the plot
fig.show()

Bar Chart

Bar charts represent data in the form of rectangular blocks; they aggregate all the data points from a dataset’s columns into individual bars. In the example below, we have used election data from the 2013 Montreal mayoral election. This dataset is provided by the Plotly library. Bar graphs represent categorical data in a graphical structure.

Python

R

Python

Python

R

import plotly.express as px
 
# using the Election dataset
df = px.data.election()

# plotting the bar chart
fig = px.bar(df, y="total", x="winner")
 
# showing the plot
fig.show()

library(plotly)
elect <- read.csv('https://raw.githubusercontent.com/plotly/datasets/master/election.csv')

fig <- plot_ly(elect,
  y = ~total,
  x = ~winner,
  name = "Winner Data",
  type = "bar"
)

fig

You can segregate the data even further by using the colour argument in bar(). In the following plot, we will segregate the win count by majority and plurality wins.

Python

R

Python

Python

R

import plotly.express as px
 
# Load the Election dataset
df = px.data.election()

# plotting the bar chart
fig = px.bar(df, y="total", x="winner",color='result')
 
# showing the plot
fig.show()

library(plotly)
elect <- read.csv('https://raw.githubusercontent.com/plotly/datasets/master/election.csv')

fig <- plot_ly(elect,
  y = ~total,
  x = ~winner,
  name = "Winner Data",
  type = "bar"
)

fig

Histogram

Histograms represent quantitative data in a graphical structure. Using histograms, we show the frequency of numerical data. We can represent the distribution of data in histograms.

Python

R

Python

Python

R

import plotly.express as px
 
# Load the Election dataset
df = px.data.election()
 
# plotting the histogram
fig = px.histogram(df, y="total", x="winner",color='result')
 
# showing the plot
fig.show()

library(plotly)
elect <- read.csv('https://raw.githubusercontent.com/plotly/datasets/master/election.csv')
fig <- plot_ly(elect,type='histogram',x=~winner,color=~result,bingroup=1,barmode='stacked')
fig

Scatter Plot

A sample scatter plot can be studied here when we plot time against the y-axis. And life expectancy in years against x, and the colour plots differentiate between continents.

Python

R

Python

Python

R

import plotly.express as px

# using the world GDP dataset
df = px.data.gapminder()

# plotting the scatter chart
fig = px.scatter(df, x="year", y="lifeExp",color='continent')

# showing the plot
fig.show()

library(plotly)
gapminder<-read.csv('https://raw.githubusercontent.com/plotly/datasets/master/gapminderDataFiveYear.csv')

fig<-plot_ly(data=gapminder,x=~year,y=~lifeExp,color=~continent)
fig

Pie Chart

The distribution of votes each contender got can be easily observed in the given pie chart.

Python

R

Python

Python

R

import plotly.express as px

# using the Election dataset
df = px.data.election()

# plotting the pie chart
fig = px.pie(df, values="total", names="winner",title='District wins Distribution')

# showing the plot
fig.show()

elect <- read.csv('https://raw.githubusercontent.com/plotly/datasets/master/election.csv')
library(plotly)
fig <- plot_ly(elect,type='pie', labels=~winner, values=~total,textinfo='label+percent',insidetextorientation='radial')
fig

Violin plot

We can see the peaks in the given plot, which is a combination of box and kernel density plots.

Python

R

Python

Python

R

import plotly.express as px

# using the tips dataset
df = px.data.election()
 
# plotting the violin chart
fig = px.violin(df, x="winner", y="total")

# showing the plot
fig.show()

library(plotly)
elect <- read.csv('https://raw.githubusercontent.com/plotly/datasets/master/election.csv')

fig <- elect %>%
plot_ly(x = ~winner,
	  y = ~total,
        split = ~winner,
        type = 'violin',
        meanline = list(
        visible = T
        )
        ) 

fig <- fig %>%
  layout(
    xaxis = list(
      title = "Winner"
    ),
    yaxis = list(
      title = "winner",
      zeroline = F
    )
  )
fig

Box plot

Skew and variance of these three groups can be seen in the given below box plot.

Python

R

Python

Python

R

import plotly.express as px
 
# using the election dataset
df = px.data.election()
 
# plotting the box plot
fig = px.box(df, x="winner", y="total")
 
# showing the plot
fig.show()

library(plotly)
elect <- read.csv('https://raw.githubusercontent.com/plotly/datasets/master/election.csv')
fig <- plot_ly(elect, y = ~total, color = ~winner, type = "box")
fig

Gantt Chart

Gantt charts are used to show the progression of an activity against time.

Python

R

Python

Python

R

import plotly.figure_factory as ff
import pandas as pd
# Data to be plotted
df = pd.DataFrame([
    dict(Task="Back-End", Start='2022-01-01', Finish='2022-04-28', Resource="Steve"),
    dict(Task="User-InterFace", Start='2022-01-01', Finish='2022-03-10', Resource="Elon"),
    dict(Task="Patch", Start='2022-04-20', Finish='2022-05-15', Resource="Mark"),
    
])
 
# Creating the plot
fig = ff.create_gantt(df)
fig.show()

install.packages('rehape2')
library(plotly)
library(reshape2)
require(ggplot2)

tasks <- c("Back-End", "User-Interface",
           "Middleware")

# Compile dataframe of task names, and respective start and end dates.
dfr <- data.frame(name = tasks,
  start.date=as.Date(c("2022-01-01","2022-01-01","2022-04-20")),
  end.date = as.Date(c("2022-4-28","2022-03-10","2022-05-15")),
  resource = c("Steve","Elon","Mark")
)

# Merge start and end dates into durations.
mdfr <- melt(dfr, measure.vars = c("start.date", "end.date"))

p <- ggplot(dfr) +
      geom_linerange(aes(y = name, 
                         xmin = start.date,
                         xmax = end.date,
                         colour = as.factor(resource)),
                     size = I(50)) +
      theme_minimal()


ggplotly(p)

Timeline plot

Timeline charts can display events in chronological order.

Python

R

Python

Python

R

import plotly.figure_factory as ff
import pandas as pd
# Data to be plotted
df = pd.DataFrame([
    dict(Task="Back-End", Start='2022-01-01', Finish='2022-03-28', Emp="Elon"),
    dict(Task="User-InterFace", Start='2022-01-01', Finish='2022-03-10', Emp="Steve"),
    dict(Task="User-InterFace", Start='2022-01-10', Finish='2022-02-25', Emp="Mark"),
    dict(Task="Patch", Start='2022-04-10', Finish='2022-05-15', Emp="Mark")   
])
 
# Creating the plot
fig = px.timeline(df,x_start='Start',x_end='Finish',y='Emp',color='Emp')
fig.show()

# Install and load the required packages
if (!require("plotly")) {
  install.packages("plotly")
}
library(plotly)

# Load the data
df <- data.frame(
  Task = c("Back-End", "User-InterFace", "User-InterFace", "Patch"),
  Start = as.Date(c("2022-01-01", "2022-01-01", "2022-01-10", "2022-04-10")),
  Finish = as.Date(c("2022-03-28", "2022-03-10", "2022-02-25", "2022-05-15")),
  Emp = c("Elon", "Steve", "Mark", "Mark")
)

# Create the plot
fig <- plot_ly(data = df, x = ~Start, xend = ~Finish, y = ~Emp, type = "scatter", mode = "lines",
               line = list(color = "blue", width = 6)) %>%
  layout(title = "Task Timeline",
         xaxis = list(title = "Timeline"),
         yaxis = list(title = "Employee"),
         showlegend = FALSE)

# Show the plot
fig

Python

Quiver Plot

These plots show vector lines as arrows. To show for example electrical potential

Python

R

Python

Python

R

import plotly.graph_objects as go
import numpy as np 

# Creating the X, Y value that will
# change the values of Z as a function
x,y = np.meshgrid(np.arange(2, 4, .3), np.arange(2, 4, .3))
u = np.cos(y)*x
v = np.sin(y)*x

fig = ff.create_quiver(x, y, u, v)
fig.show()

library(plotly)

df <- expand.grid(x = 1:20, y=1:20)
df$angle <- runif(100, 0, 2*3.14)
df$speed <- runif(100, 0, sqrt(df$x))

p <- ggplot(df, aes(x, y)) +
  geom_point() +
  geom_spoke(aes(angle = angle, radius = speed))

fig <- ggplotly(p)

fig

Heatmap

Heatmap visualisations show represent the data magnitude in the form of colour. More magnitude means darker colours.

Python

R

Python

Python

R

import plotly.graph_objects as go
 
feature_x = np.arange(0, 100, 2)
feature_y = np.arange(0, 100, 3)

# Creating 2-D grid of features
[X, Y] = np.meshgrid(feature_x, feature_y)

Z = np.cos(X / 2) + np.sin(Y / 4)

# plotting the figure
fig = go.Figure(data =
     go.Heatmap(x = feature_x, y = feature_y, z = Z,))

fig.show()

library(plotly)
fig <- plot_ly(z=~volcano,type = "heatmap") %>%
    layout(margin = list(l=120))

fig