Plotly

Plotly is a powerful and flexible library for creating interactive, publication-quality graphs in Jupyter Labs. It integrates seamlessly with Python and Jupyter, enabling data scientists and analysts to visualize complex datasets easily.
Author

Benedict Thekkel

1. Overview

  • Plotly is an open-source graphing library that makes interactive, publication-quality graphs.
  • It is built on top of D3.js and WebGL.
  • Integrates seamlessly with Jupyter Labs and supports inline rendering.
  • It has two main interfaces:
    • Plotly Express: High-level interface for quick, concise plotting.
    • Graph Objects: Low-level, flexible interface for detailed customization.

2. Installing Plotly for Jupyter Labs

To get started with Plotly in Jupyter Labs, you need to install the following:

Install Plotly Library

pip install plotly

Install JupyterLab Extensions

To render Plotly figures within JupyterLab, you need to install the following extensions:

# JupyterLab 3.x and above (Extensions are auto-managed)
pip install jupyterlab "ipywidgets>=7.5"

# For older JupyterLab versions, you might need:
jupyter labextension install jupyterlab-plotly
jupyter labextension install @jupyter-widgets/jupyterlab-manager

Check Installation


import plotly
print(plotly.__version__)  # Should be 5.x or above

Ensure you have Plotly version 5.x or above for the latest features.

import plotly
print(plotly.__version__)  # Should be 5.x or above
6.0.0

3. Importing Plotly

import plotly.express as px
import plotly.graph_objects as go
  • Plotly Express (px): High-level, easy-to-use interface for rapid plotting.
  • Graph Objects (go): Low-level, flexible API for detailed customization.
import plotly.express as px
import plotly.graph_objects as go

4. Plotly Express Basics

4.1. Line Chart

import plotly.express as px
import numpy as np
import pandas as pd

# Sample Data
x = np.linspace(0, 10, 100)
y = np.sin(x)

df = pd.DataFrame({'x': x, 'y': y})

# Line Plot
fig = px.line(df, x='x', y='y', title='Sine Wave')
fig.show()

### **4.2. Scatter Plot**

```python
fig = px.scatter(df, x='x', y='y', title='Scatter Plot of Sine Wave')
fig.show()

4.3. Bar Chart

data = {'Fruits': ['Apple', 'Banana', 'Cherry', 'Date'],
        'Quantity': [10, 20, 15, 5]}

df = pd.DataFrame(data)

fig = px.bar(df, x='Fruits', y='Quantity', title='Fruit Quantity')
fig.show()

4.4. Histogram

fig = px.histogram(df, x='Quantity', title='Histogram of Quantity')
fig.show()

4.5. Pie Chart

fig = px.pie(df, names='Fruits', values='Quantity', title='Fruit Distribution')
fig.show()
import plotly.express as px
import numpy as np
import pandas as pd
import plotly.offline as pyo
from IPython.display import HTML, display
# Sample Data
x = np.linspace(0, 10, 100)
y = np.sin(x)

df = pd.DataFrame({'x': x, 'y': y})
# Turn off the menu bar
config = {
    'displayModeBar': False  # Disable the mode bar (menu bar)
}

# Line Plot
fig = px.line(df, x='x', y='y', title='Sine Wave')
# fig.show(config=config)
fig_html = fig.to_html(full_html=False, config=config)
display(HTML(fig_html))

fig = px.scatter(df, x='x', y='y', title='Scatter Plot of Sine Wave')
# fig.show(config=config)
fig_html = fig.to_html(full_html=False, config=config)
display(HTML(fig_html))
data = {'Fruits': ['Apple', 'Banana', 'Cherry', 'Date'],
        'Quantity': [10, 20, 15, 5]}

df = pd.DataFrame(data)

fig = px.bar(df, x='Fruits', y='Quantity', title='Fruit Quantity')
fig_html = fig.to_html(full_html=False, config=config)
display(HTML(fig_html))


fig = px.histogram(df, x='Quantity', title='Histogram of Quantity')
fig_html = fig.to_html(full_html=False, config=config)
display(HTML(fig_html))


fig = px.pie(df, names='Fruits', values='Quantity', title='Fruit Distribution')
fig_html = fig.to_html(full_html=False, config=config)
display(HTML(fig_html))

5. Advanced Plotly Express Features

5.1. Facet Plots

df = px.data.tips()

fig = px.scatter(df, x="total_bill", y="tip", color="sex",
                 facet_col="time", facet_row="smoker",
                 title="Facet Plot of Tips")
fig.show()

5.2. Animations

df = px.data.gapminder()

fig = px.scatter(df, x="gdpPercap", y="lifeExp", animation_frame="year", 
                 animation_group="country", size="pop", color="continent",
                 hover_name="country", log_x=True, size_max=55, 
                 range_x=[100,100000], range_y=[25,90],
                 title="Gapminder Animation")
fig.show()

5.3. 3D Scatter Plot

fig = px.scatter_3d(df, x='gdpPercap', y='lifeExp', z='pop', 
                    color='continent', size='pop', 
                    hover_name='country', log_x=True, 
                    title="3D Scatter Plot of Gapminder Data")
fig.show()

5.4. Geographic Plots

fig = px.choropleth(df, locations="iso_alpha",
                    color="lifeExp",
                    hover_name="country",
                    animation_frame="year",
                    title="World Life Expectancy Over Time")
fig.show()
df = px.data.tips()

fig = px.scatter(df, x="total_bill", y="tip", color="sex",
                 facet_col="time", facet_row="smoker",
                 title="Facet Plot of Tips")
fig.show()

df = px.data.gapminder()

fig = px.scatter(df, x="gdpPercap", y="lifeExp", animation_frame="year", 
                 animation_group="country", size="pop", color="continent",
                 hover_name="country", log_x=True, size_max=55, 
                 range_x=[100,10000], range_y=[25,90],
                 title="Gapminder Animation")
fig.show()

6. Plotly Graph Objects (Advanced Customization)

6.1. Line Chart

fig = go.Figure()

fig.add_trace(go.Scatter(x=x, y=y, mode='lines+markers', name='Sine Wave'))
fig.update_layout(title='Sine Wave', xaxis_title='X', yaxis_title='Y')
fig.show()

6.2. Subplots

from plotly.subplots import make_subplots

fig = make_subplots(rows=2, cols=2)

fig.add_trace(go.Scatter(x=x, y=y, mode='lines', name='Line Plot'), row=1, col=1)
fig.add_trace(go.Scatter(x=x, y=np.cos(x), mode='markers', name='Scatter Plot'), row=1, col=2)
fig.add_trace(go.Bar(x=['A', 'B', 'C'], y=[5, 10, 15], name='Bar Chart'), row=2, col=1)
fig.add_trace(go.Pie(labels=['Apple', 'Banana', 'Cherry'], values=[10, 20, 30], name='Pie Chart'), row=2, col=2)

fig.update_layout(title='Multiple Plots')
fig.show()

6.3. Annotations and Shapes

fig.update_layout(
    annotations=[dict(x=2, y=0, xref="x", yref="y", text="Annotation",
                      showarrow=True, arrowhead=2)],
    shapes=[dict(type="line", x0=2, y0=-1, x1=2, y1=1, 
                 line=dict(color="RoyalBlue", width=2))]
)
fig.show()
fig = go.Figure()

fig.add_trace(go.Scatter(x=x, y=y, mode='lines+markers', name='Sine Wave'))
fig.update_layout(title='Sine Wave', xaxis_title='X', yaxis_title='Y')
fig.show()

from plotly.subplots import make_subplots

fig = make_subplots(rows=2, cols=2)

fig.add_trace(go.Scatter(x=x, y=y, mode='lines', name='Line Plot'), row=1, col=1)
fig.add_trace(go.Scatter(x=x, y=np.cos(x), mode='markers', name='Scatter Plot'), row=1, col=2)
fig.add_trace(go.Bar(x=['A', 'B', 'C'], y=[5, 10, 15], name='Bar Chart'), row=2, col=1)
# fig.add_trace(go.Pie(labels=['Apple', 'Banana', 'Cherry'], values=[10, 20, 30], name='Pie Chart'), row=2, col=2)

fig.update_layout(title='Multiple Plots')
fig.show()

fig.update_layout(
    annotations=[dict(x=2, y=0, xref="x", yref="y", text="Annotation",
                      showarrow=True, arrowhead=2)],
    shapes=[dict(type="line", x0=2, y0=-1, x1=2, y1=1, 
                 line=dict(color="RoyalBlue", width=2))]
)
fig.show()

7. Saving and Exporting

7.1. Static Images

fig.write_image("plot.png")   # PNG format
fig.write_image("plot.pdf")   # PDF format

Dependencies:

pip install -U kaleido  # Required for static image export

7.2. Interactive HTML

fig.write_html("plot.html", auto_open=True)

8. Integration with Jupyter Widgets

import ipywidgets as widgets

def update_plot(change):
    new_data = np.sin(x + change['new'])
    fig = px.line(x=x, y=new_data, title='Interactive Sine Wave')
    fig.show()

slider = widgets.FloatSlider(value=0, min=0, max=10, step=0.1, description='Phase')
slider.observe(update_plot, names='value')
display(slider)

9. Best Practices and Tips

  • Use Plotly Express for quick prototyping and rapid visualization.
  • Use Graph Objects for advanced customizations and complex layouts.
  • Optimize Performance:
    • Use scattergl for large datasets (WebGL acceleration).
    • Reduce DOM size by limiting data points or using aggregation.
  • JupyterLab Integration:
    • Ensure JupyterLab extensions are up to date for optimal rendering.
    • Use fig.show(renderer='notebook') for inline rendering in some cases.

10. Common Issues and Solutions

10.1. Graphs Not Displaying in Jupyter Lab

  • Ensure JupyterLab extensions are installed and updated.
  • Clear browser cache or try a different browser.
  • Use %matplotlib inline to avoid conflicts with Matplotlib.

10.2. Exporting as Static Images Not Working

  • Install kaleido for image exporting:

    pip install -U kaleido

10.3. Performance Lag with Large Datasets

  • Use scattergl for WebGL acceleration.
  • Consider down-sampling or aggregating data.
Back to top