Visualization Libraries and Tools

TODO: AS WE WILL REPLACE FOOTNOTES WITH PROPER CITATIOPNS FOOTNOTS CAN NOT BE BEHIND A DO THE MUST BEE BEFORE THE DOT.

TODO: any citation that is not listed as citation in the main text will be deleted automatically. so make sure they are used if you add them

TODO: A CITATION CAN NOT BE BEHIND A cod, it must be before and used in a proper centence

In this section we will showcase a number of visualization tools that makes it easy to display data and also allow for data analysis while using graphical means.

These tools are summarized in the follwing table where we list the name of the tool a link to the documentation, a link to a gallarey, as well as comments about functionality of the library.

Name	Refernce	Link	Documentation	Gallery	Features
pandas	¹	TBD	TBD	TBD	data frames, uses matplotlib
matplotlib	TBD	TBD	TBD	Gallery	Features

Installation

In order for simplifying the instalation we will provide here the instalation commands for all libraries we describe. If you do not want to install the one or the other library, please exclude the instalation. As we focus on Python tools, most libraries can be installed just with pip. However some do require other support packages. We have integrated also them into our quick instalation guide.

At this time we only provde instalations for Unbuntu. Please provide us with updates to the instalations for other OS.

In case you use Windows, we assume you use gitbash so that the differences between Linux and Windows are minimal We Will be using a python virtualenv on all systems as to avoid compatibility issues that may arise unexpectetly. Please refer to this documentation if you are not familiar with ENV3 based installation:

https://cybertraining-dsc.github.io/docs/tutorial/reu/python/

$ pip install numpy
$ pip install matplotlib
$ sudo apt install graphviz -y # Install the grapghviz binary
$ pip install graphviz
$ pip install seaborn
$ pip install plotnine         # Install ggplot
$ pip install bokeh
$ pip install pygal
$ pip install plotly
$ pip install altair vega_datasets   # install altair,
                                     #  but also test data
$ pip install pyglet           # used for geoplotlib (is this not in requirements?)
$ pip install geoplotlib
$ pip install gleam
$ pip install missingno
$ pip install bqplot
$ pip install pandas
$ pip install jupyterlab
$ pip install notebook
$ pip install ipywidgets
$ pip install tqdm

GraphViz

graphviz is an open source python mod that is used to create graph visualizations as a way of representing structural information as graphs and networks ².

Gallery

https://graphviz.org/gallery/

matplotlib

Matplotlib is a library in Python for creating static (still figures), animated (moving pictures), and interactive vizualizations such as charts, graphs and scatter plots. It is one of the oldest and most popular plotting libraries in Python.

Installation

Gallery

https://matplotlib.org/stable/gallery/index.html

Matplotlib Stacked Bar Chart Example

³ PLEASE INVENT YOUR OWN

import matplotlib.pyplot as plt


labels = ['G1', 'G2', 'G3', 'G4', 'G5']
men_means = [20, 35, 30, 35, 27]
women_means = [25, 32, 34, 20, 25]
men_std = [2, 3, 4, 1, 2]
women_std = [3, 5, 2, 3, 3]
width = 0.35       # the width of the bars: can also be len(x) sequence

fig, ax = plt.subplots()

ax.bar(labels, men_means, width, yerr=men_std, label='Men')
ax.bar(labels, women_means, width, yerr=women_std, bottom=men_means,
       label='Women')

ax.set_ylabel('Scores')
ax.set_title('Scores by group and gender')
ax.legend()

plt.show()

seaborn

Seaborn is a data vizualization tool in Python based on matplotlib. Seaborn is used for making statistical graphics, Helping you to explore and understand your data with a visual representation. It allows you to focus on the meaning of your plots rather than on the details of drawing them.

Gallery

https://seaborn.pydata.org/examples/index.html

Seaborn Scatterplot Example

PLEASE INVENT YOUR OWN ⁴

import seaborn as sns
import matplotlib.pyplot as plt
sns.set_theme(style="whitegrid")

# Load the example diamonds dataset
diamonds = sns.load_dataset("diamonds")

# Draw a scatter plot while assigning point colors and sizes to different
# variables in the dataset
f, ax = plt.subplots(figsize=(6.5, 6.5))
sns.despine(f, left=True, bottom=True)
clarity_ranking = ["I1", "SI2", "SI1", "VS2", "VS1", "VVS2", "VVS1", "IF"]
sns.scatterplot(x="carat", y="price",
                hue="clarity", size="depth",
                palette="ch:r=-.2,d=.3_r",
                hue_order=clarity_ranking,
                sizes=(1, 8), linewidth=0,
                data=diamonds, ax=ax)

ggplot

ggplot is run in plotnine, a grammar of graphics package in Python. It is based on the library written in R.

GGplot bar chart

PLEASE INVENT YOUR OWN ⁵

from plotnine.data import mpg
from plotnine import ggplot, aes, geom_bar

ggplot(mpg) + aes(x="class") + geom_bar()

bokeh

bokeh is a Python library used mostly for creating interactive vizualization sets. vizualizations created in bakeh are powered by JavaScript without the user having to write any JavaScript themselves. Vizualizations created in bokeh can range from simple line plots to large and complex dashboards containing multiple datasets.

Gallery

https://docs.bokeh.org/en/latest/docs/gallery.html#gallery

Bokeh Stacked Bar Chart Example

DID YOU INVENT THIS OR IS CITATION MISSING

from bokeh.plotting import figure, show

labels = ['G1', 'G2', 'G3', 'G4', 'G5']
genders = ["Men", "Women"]
colors = ["#c9d9d3", "#718dbf"]

data = {'labels' : labels,
        'Men'   : [20, 35, 30, 35, 27],
        'Women'   : [25, 32, 34, 20, 25]}

p = figure(x_range=labels, height=250, title="Scores by group and gender",
           toolbar_location=None, tools="hover", tooltips="$name @labels: @$name")

p.vbar_stack(genders, x='labels', width=0.9, color=colors, source=data,
             legend_label=genders)

p.y_range.start = 0
p.x_range.range_padding = 0.1
p.xgrid.grid_line_color = None
p.axis.minor_tick_line_color = None
p.outline_line_color = None
p.legend.location = "top_left"
p.legend.orientation = "horizontal"

show(p)

altair

Altair is a declarative statistical vizualization library for Python ⁶.

The above line of code only installs a partial version of altair. You can find instructions on how to install the master version of Altair here:

https://altair-viz.github.io/getting_started/installation.html

Gallery

https://altair-viz.github.io/gallery/index.html

Altair Bar Chart Example

This example uses pandas in conjunction with altair, pandas installation is referred to later on this page.

PLEASE INVENT YOUR OWN ⁷

import altair as alt
import pandas as pd

source = pd.DataFrame({
    'a': ['A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I'],
    'b': [28, 55, 43, 91, 81, 53, 19, 87, 52]
})

alt.Chart(source).mark_bar().encode(
    x='a',
    y='b'
)

pygal

pygal is a dynamic Scalable Vector Graphics (SVG) library in Python.

Gallery

http://www.pygal.org/en/latest/index.html

plotly

plotly is a graphing library in Python that is used to make interactive, publication quality graphs. It is built on top of plotly's Javascript library and supports over 40 chart types.

Gallery (basic charts)

https://plotly.com/python/basic-charts/

Plotly Stacked Bar Chart Example

IS THIS INVENTED BY YOU?

import plotly.graph_objects as go
labels = ['G1', 'G2', 'G3', 'G4', 'G5']

fig = go.Figure(data=[
    go.Bar(name='Women', x=labels, y=[25, 32, 34, 20, 25]),
    go.Bar(name='Men', x=labels, y=[20, 35, 30, 35, 27])
])
# Change the bar mode
fig.update_layout(barmode='stack')
fig.show()

geoplotlib

geoplotlib is a Python toolkit for creating vizualization of geographical data. It can be used to make maps, spatial graphs, etc.

Examples (dot density and spatial graph)

https://www.pluralsight.com/guides/building-geoplots-with-geoplotlib

geoplotlib Dot Density Map Example

gleam

Gleam allows a user to make interactive web interface versions of plots and graphs. A user can control a specified number of inputs and you can use any sort of Python graphing library to plot those inputs. Gleam provides the web interface that allows anyone to control the data in real time.

Gleam Interactive Scatter Plot Example

PLEASE INVET YOUR OWN ⁸

from wtforms import fields
from ggplot import *
from gleam import Page, panels

class ScatterInput(panels.Inputs):
    title = fields.StringField(label="Title of plot:")
    yvar = fields.SelectField(label="Y axis",
                              choices=[("beef", "Beef"),
                                       ("pork", "Pork")])
    smoother = fields.BooleanField(label="Smoothing Curve")

class ScatterPlot(panels.Plot):
    name = "Scatter"

    def plot(self, inputs):
        p = ggplot(meat, aes(x='date', y=inputs.yvar))
        if inputs.smoother:
            p = p + stat_smooth(color="blue")
        p = p + geom_point() + ggtitle(inputs.title)
        return p

class ScatterPage(Page):
    input = ScatterInput()
    output = ScatterPlot()

ScatterPage.run()

missingno

missingno provides a series of visualizations to help users understand the presence and distribution of missing data within the pandas datafram, which we will discuss later ⁹. From the plots created using missingno, users can identify where missing values occur, if they are correlated, etc.

missingno Bar Example

This example uses data from the NYPD Motor Vehicle Collisions Dataset ¹⁰.

¹¹

import pandas as pd
import missingno as msno
%matplotlib inline

# importing the dataset from 
collisions = pd.read_csv("https://raw.githubusercontent.com/ResidentMario/missingno-data/master/nyc_collision_factors.csv")

msno.bar(collisions.sample(1000))

bqplot

bqplot is a 2-D interactive visualization library in the Jupyter Notebook. In bqplot, all components are interactive, allowing the user to integrate visualization s with other Jupyter interactive widgets to create integrated GUIs with a few lines of Python code ¹².

Gallery

https://bqplot.readthedocs.io/en/latest/usage.html#examples

Bqplot Line Chart Example

import numpy as np
from bqplot import pyplot as plt

plt.figure(1, title='Line Chart')
np.random.seed(0)
n = 200
x = np.linspace(0.0, 10.0, n)
y = np.cumsum(np.random.randn(n))
plt.plot(x, y)
plt.show()

¹³

pandas

pandas is an open source data analysis and manipulation tool. It is build on top of the Python language.

Pandas Bar Chart Example using Altair

PLEASE INVENT YOUR OWN ⁷

import altair as alt
import pandas as pd

source = pd.DataFrame({
    'a': ['A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I'],
    'b': [28, 55, 43, 91, 81, 53, 19, 87, 52]
})

alt.Chart(source).mark_bar().encode(
    x='a',
    y='b'
)

ipywidgets

ipywidgets are HTML-interactive widgets used for Jupyter notebooks to help users visualize changes in data.

Installation

Do we have to configure jupyter? jupyter-lab

jupyter

Jupyter consists of two web-based interfaces: JupyterLab and Jupyter Notebook. JupyterLab is a development environment for various jupyter notebooks code and data. Jupyter Notebook is an open-source application that allows users to create and share documents that contain live code, equations, vizualization and narrative text ¹⁴.

TBD: DIFFERENCE NOT CLEAR

Running JupyterLab

$ jupyter-lab

Running Jupyter Notebook

$ jupyter notebook

tqdm

tqdm is a Python library that is used for creating progress meters/bars.

Installation

There may be a special version and use for jupyter vs regular

tqdm simple example

PLAEASE INVENT YOUR OWN ¹⁵

import tqdm
for i in tqdm.trange(int(1e8)):
    pass

Assignments

Fix the citation issues for codes
make sure you define an uniform example tha you use in all tools
Please check in all working code intot the repository with nice nameing scheme such as pandas_barchart.py Use the url to that code in your documentation above the code.
Add a section ## Image Manipulation

Resources

Footnotes

Missing ↩
https://graphviz.org/ ↩
https://matplotlib.org/stable/gallery/lines_bars_and_markers/bar_stacked.html#sphx-glr-gallery-lines-bars-and-markers-bar-stacked-py ↩
https://seaborn.pydata.org/examples/different_scatter_variables.html ↩
https://realpython.com/ggplot-python/#plotting-data-using-python-and-ggplot ↩
https://altair-viz.github.io/ ↩
https://altair-viz.github.io/gallery/simple_bar_chart.html ↩ ↩²
https://github.com/dgrtwo/gleam ↩
https://towardsdatascience.com/using-the-missingno-python-library-to-identify-and-visualise-missing-data-prior-to-machine-learning-34c8c5b5f009 ↩
https://data.cityofnewyork.us/Public-Safety/Motor-Vehicle-Collisions-Crashes/h9gi-nx95 ↩
https://github.com/ResidentMario/missingno ↩
https://github.com/bqplot/bqplot ↩
https://bqplot.readthedocs.io/en/latest/usage.html#examples ↩
https://jupyter.org/ ↩
https://github.com/tqdm/tqdm/blob/master/examples/simple_examples.py ↩

Files

LibrariesandTools.md

Latest commit

History

LibrariesandTools.md

File metadata and controls

Visualization Libraries and Tools

Installation

GraphViz

Gallery

matplotlib

Installation

Gallery

Matplotlib Stacked Bar Chart Example

seaborn

Gallery

Seaborn Scatterplot Example

ggplot

GGplot bar chart

bokeh

Gallery

Bokeh Stacked Bar Chart Example

altair

Gallery

Altair Bar Chart Example

pygal

Gallery

plotly

Gallery (basic charts)

Plotly Stacked Bar Chart Example

geoplotlib

Examples (dot density and spatial graph)

geoplotlib Dot Density Map Example

gleam

Gleam Interactive Scatter Plot Example

missingno

missingno Bar Example

bqplot

Gallery

Bqplot Line Chart Example

pandas

Pandas Bar Chart Example using Altair

ipywidgets

Installation

jupyter

Running JupyterLab

Running Jupyter Notebook

tqdm

Installation

tqdm simple example

Assignments

Resources

Footnotes