Histograms in Plotly

Histograms are an essential tool for visualizing the distribution of a dataset. In this tutorial, we will walk through the steps to create and customize histograms using the Plotly library.

### Basic Histogram

To get started, let's create a simple histogram from a dataset.


import plotly.express as px
import numpy as np

# Generate random data
np.random.seed(0)
data = np.random.randn(1000)

# Create a basic histogram
fig = px.histogram(data, title='Basic Histogram')

# Show the figure
fig.show()
- **`px.histogram(data, title='Basic Histogram')`** creates a histogram

### Customizing Histogram Bins

You can customize the number of bins and the bin width to better represent your data.


import plotly.express as px
import numpy as np

# Generate random data
np.random.seed(0)
data = np.random.randn(1000)

# Create a histogram with custom bin width
fig = px.histogram(data, nbins=100, title='Histogram with Custom Bins')

# Show the figure
fig.show()
- **`nbins=100`** specifies the number of bins.

### Customizing Axis Labels and Titles

You can add and customize labels and titles to make your plot more informative.


import plotly.express as px
import numpy as np

# Generate random data
np.random.seed(0)
data = np.random.randn(500)

# Create a histogram with labels and titles
fig = px.histogram(
    data, 
    nbins=30, 
    title='Histogram with Custom Labels',
)
fig.update_layout(xaxis_title='Value', yaxis_title='Frequency')

# Show the figure
fig.show()
- **`title='Histogram with Custom Labels'`** sets the plot title
- **`fig.update_layout(xaxis_title='Value', yaxis_title='Frequency')`** sets axis labels

### Creating Overlaid Histograms

You can overlay multiple histograms for comparative analysis.


import plotly.graph_objects as go

# Generate two sets of random data
np.random.seed(0)
data1 = np.random.randn(500)
data2 = np.random.randn(500) + 2

# Create histograms for both datasets
fig = go.Figure()
fig.add_trace(go.Histogram(x=data1, nbinsx=30, name='Dataset 1', opacity=0.75))
fig.add_trace(go.Histogram(x=data2, nbinsx=30, name='Dataset 2', opacity=0.75))

# Overlay both histograms by setting the bar mode
fig.update_layout(
    barmode='overlay',
    title='Overlaid Histograms',
    xaxis_title='Value',
    yaxis_title='Frequency'
)

# Show the figure
fig.show()
- **`go.Figure()`** creates a new figure.
- **`go.Histogram(x=data1, nbinsx=30, name='Dataset 1', opacity=0.75)`** defines the first histogram with 30 bins and 75% opacity.
- **`barmode='overlay'`** overlays the histograms.

### Creating Stacked Histograms

You can also stack multiple histograms.


import plotly.graph_objects as go

# Generate two sets of random data
np.random.seed(0)
data1 = np.random.randn(500)
data2 = np.random.randn(500) + 2

# Create histograms for both datasets
fig = go.Figure()
fig.add_trace(go.Histogram(x=data1, nbinsx=30, name='Dataset 1'))
fig.add_trace(go.Histogram(x=data2, nbinsx=30, name='Dataset 2'))

# Stack the histograms by setting the bar mode
fig.update_layout(
    barmode='stack',
    title='Stacked Histograms',
    xaxis_title='Value',
    yaxis_title='Frequency'
)

# Show the figure
fig.show()
- **`barmode='stack'`** stacks the histograms.

### Customizing Histogram Colors

You can customize the colors of your histograms for better aesthetics.


import plotly.graph_objects as go

# Generate two sets of random data
np.random.seed(0)
data1 = np.random.randn(500)
data2 = np.random.randn(500) + 2

# Create histograms for both datasets with custom colors
fig = go.Figure()
fig.add_trace(go.Histogram(x=data1, nbinsx=30, name='Dataset 1', marker_color='blue'))
fig.add_trace(go.Histogram(x=data2, nbinsx=30, name='Dataset 2', marker_color='orange'))

# Overlay both histograms by setting the bar mode
fig.update_layout(
    barmode='overlay',
    title='Histograms with Custom Colors',
    xaxis_title='Value',
    yaxis_title='Frequency'
)

# Show the figure
fig.show()
- **`marker_color='blue'`** and **`marker_color='orange'`** set custom colors for the histograms.

### Histogram with Density Estimation using `distplot`

The `distplot` function from Plotly's `figure_factory` module can be used to create a histogram with a density curve for visualizing the probability density function (PDF).


import plotly.figure_factory as ff
import numpy as np

# Generate random data
np.random.seed(0)
data = np.random.randn(1000)

# Create a distplot
fig = ff.create_distplot([data], group_labels=['Data'], bin_size=0.2, show_hist=True, show_curve=True)

# Update layout
fig.update_layout(
    title='Histogram with Density Estimation',
    xaxis_title='Value',
    yaxis_title='Density'
)

# Show the figure
fig.show()
- **`ff.create_distplot([data], group_labels=['Data'], bin_size=0.2, show_hist=True, show_curve=True)`** creates a histogram with a density curve.
  - **`bin_size=0.2`** defines the bin size.
  - **`show_hist=True`** shows the histogram.
  - **`show_curve=True`** shows the PDF curve.


### Conclusion

Using Plotly, you can create interactive and customizable histograms to visualize the distribution of your data effectively. Plotly offers convenient functions for overlaying or stacking histograms, customizing colors, or adding density curves.

You may also like