Outline¶
- Why Interactive Visualization?
- Plotly Express - Intro
- Basic Visualizations
- Improving a Plot - One Component at a time
- Building a Story - with one line of Code
Why Interactive Visualization?¶
In [1]:
import seaborn as sns
import matplotlib.pyplot as plt
crashes = sns.load_dataset("car_crashes")
sns.set(rc={'figure.figsize':(11.7,8.27)})
Static Visualization - Find the Values¶
In [2]:
sns.barplot(x = 'speeding', y = 'abbrev', data = crashes);
Interactive Visualization - Just, Hover over!¶
In [3]:
import plotly_express as px
px.bar(data_frame = crashes, x = 'speeding', y = 'abbrev', color = 'abbrev', orientation= "h")
Interactive over Static¶
- Intuitive - More Readable
- Less Code, More Stories/Insights
- Customizable - Pan, Zoom, Select
- This is 2019!!!
Plotly Express 
Plotly Express¶
- High-level Python visualization library
- Wrapper around
plotly.py - Plotly Express is totally free
- Right now, Only available for Python
- A simple syntax for complex charts
- Inspired by
Seabornandggplot2(R) - Consisten & Easy-to-learn API
- Offline-mode by Default
Installation¶
In [4]:
# !! pip3 install plotly_express # Python3
Loading¶
In [5]:
import plotly_express as px
Dataset¶
In [6]:
gapminder = px.data.gapminder()
gapminder.tail(4)
Out[6]:
Simple Visualizations¶
Bar¶
In [7]:
px.bar(data_frame= gapminder[gapminder.year==2007], x = 'country', y = 'pop')
Histogram¶
In [8]:
px.histogram(data_frame= gapminder, x = 'pop')
Box plot - Normal¶
In [9]:
px.box(data_frame= gapminder, y = 'lifeExp')
Box Plot - Colored / Grouped¶
In [10]:
px.box(data_frame= gapminder, y = 'lifeExp', color = 'continent')
Violin plot¶
In [11]:
px.violin(data_frame= gapminder, y = 'lifeExp', color = 'continent', box=True)
Line Graph¶
In [12]:
px.line(gapminder.query("country == ['India','China','United States']"), 'year','pop', color = 'country')
Improving a Plot - One Component at a Time¶
Scatter Plot¶
In [13]:
px.scatter(data_frame= gapminder, x = 'gdpPercap', y = 'lifeExp')
Scatter Plot with Box and Histogram¶
In [14]:
px.scatter(data_frame= gapminder, x = 'gdpPercap', y = 'lifeExp', marginal_y='box', marginal_x='histogram')
Scatter Plot with Trend Line¶
In [15]:
px.scatter(data_frame= gapminder, x = 'gdpPercap', y = 'lifeExp', trendline='ols')
Scatter Plot with Country Values (Hover)¶
In [16]:
px.scatter(data_frame= gapminder, x = 'gdpPercap', y = 'lifeExp', hover_name = 'country', trendline= 'ols')
Scatter Plot with Country Values (Hover) & Color - Transformed Axis¶
In [17]:
px.scatter(data_frame= gapminder, x = 'gdpPercap', y = 'lifeExp', hover_name = 'country', color = 'country', log_x=True)
In [18]:
gapminder7 = gapminder[gapminder.year==2007]
Scatter Plot¶
In [19]:
px.scatter(data_frame= gapminder7, x = 'gdpPercap', y = 'lifeExp', hover_name = 'country',
log_x=True)
Scatter Plot - Transformed Axis & Facet - with Trendline¶
In [20]:
px.scatter(data_frame= gapminder7, x = 'gdpPercap', y = 'lifeExp', hover_name = 'country',
log_x=True, facet_col="continent", trendline = 'ols')
Bubble Chart¶
In [21]:
px.scatter(data_frame= gapminder7, x = 'gdpPercap', y = 'lifeExp', hover_name = 'country',
log_x=True, #facet_col="continent",
color = 'continent', size = 'pop')
Bubble Chart with Facet by Continent¶
In [22]:
px.scatter(data_frame= gapminder7, x = 'gdpPercap', y = 'lifeExp', hover_name = 'country',
log_x=True, facet_col="continent",
color = 'continent', size = 'pop')
Scatter Plot with Country Values (Hover) & Color¶
In [23]:
px.scatter(data_frame= gapminder7, x = 'gdpPercap', y = 'lifeExp', hover_name = 'country', color = 'country')
In [24]:
px.scatter(data_frame= gapminder7, x = 'gdpPercap', y = 'lifeExp', hover_name = 'country', color = 'country',
size = 'pop')
The Big Picture - Timeline Animation¶
In [25]:
px.scatter(px.data.gapminder(), x="gdpPercap", y="lifeExp", animation_frame="year", animation_group="country",
size="pop", color="country", hover_name="country",
log_x = True,
size_max=45, range_x=[100,100000], range_y=[25,90])