Heatmap in Plotly

Last Updated: 08 Nov 2025

Heatmaps show data density or intensity using color gradients — perfect for correlation matrices, time vs category sales, website clicks, or sensor grid data.

Hinglish Tip: Heatmap = "jahan zyada value, wahan garm color" — jaise weather map! Dark red = high, blue = low.

Basic Heatmap

import plotly.express as px
import numpy as np

# Sample 5x5 matrix
data = np.random.rand(5, 5)

fig = px.imshow(data, title="Random Heatmap")
fig.show()

Key Parameters in `px.imshow()` (Heatmap)

Parameter	Values / Examples	Purpose
text_auto	True, '.2f'	Show values on cells
color_continuous_scale	'Viridis', 'Reds', 'Blues'	Color gradient
zmin, zmax	0, 1	Fix color scale range
labels	'x': 'Day', 'y': 'Hour'	Axis labels
x, y	List of labels	Row/column names
aspect	'auto', 'equal'	Cell shape
title	String	Chart title
color_continuous_midpoint	0	For diverging scales

Full Example: Website Traffic (Hour vs Day)

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

# Generate sample traffic data
np.random.seed(42)
days = ['Mon', 'Tue', 'Wed', 'Thu', 'Fri', 'Sat', 'Sun']
hours = [f'{h:02d}:00' for h in range(24)]

traffic = np.random.randint(50, 500, size=(len(hours), len(days)))

df = pd.DataFrame(traffic, index=hours, columns=days)

fig = px.imshow(
    df,
    text_auto=True,
    color_continuous_scale='YlOrRd',  # Yellow to Red
    title='Website Traffic: Hour vs Day',
    labels=dict(x='Day', y='Hour', color='Visitors'),
    zmin=0,
    zmax=500,
    aspect='auto'
)

fig.update_layout(
    font=dict(size=10),
    coloraxis_colorbar=dict(
        title="Visitors",
        tickvals=[0, 250, 500],
        ticktext=['Low', 'Medium', 'High']
    )
)

fig.show()

Hinglish Tip: text_auto=True → har cell mein number dikhega — no guesswork!

Correlation Heatmap

import plotly.express as px
import pandas as pd
import seaborn as sns

# Sample dataset
df = sns.load_dataset('iris').select_dtypes('number')

corr = df.corr()

fig = px.imshow(
    corr,
    text_auto='.2f',
    color_continuous_scale='RdBu_r',  # Red-Blue diverging
    color_continuous_midpoint=0,
    title='Iris Dataset: Feature Correlation',
    labels=dict(color='Correlation'),
    zmin=-1, zmax=1
)

fig.update_layout(
    xaxis_title="", yaxis_title="",
    coloraxis_colorbar=dict(
        title="Corr",
        tickvals=[-1, -0.5, 0, 0.5, 1],
        ticktext=['-1', '-0.5', '0', '0.5', '+1']
    )
)

fig.show()

Hinglish Tip: RdBu_r → red = negative, blue = positive — ML mein correlation samajhne ke liye best!

Annotated Heatmap with Custom Labels

fig = px.imshow(
    df.head(10),  # First 10 hours
    text_auto='.0f',
    color_continuous_scale='Greens',
    title='Top 10 Hours Traffic',
    labels=dict(color='Users')
)

fig.update_xaxes(side="top")  # Labels on top
fig.show()

Common Mistakes to Avoid

Mistake	Fix
No text on cells	Use text_auto=True or '.1f'
Color scale auto → misleading	Fix with zmin, zmax
Too many rows/cols → unreadable	Filter top N or use df.head(20)
Wrong color scheme	Use Viridis (sequential), RdBu_r (diverging)
Can't export	Install kaleido

← Previous Next →