Data Visualisation

1Matplotlib fundamentals

Matplotlib is the foundation every Python chart is built on. The professional pattern is fig, ax = plt.subplots() — you get a figure (the canvas) and an axes (one plot) you control precisely.

import matplotlib.pyplot as plt

regions = ['North', 'East', 'West', 'South']
sales   = [5300, 4100, 3800, 2940]

fig, ax = plt.subplots(figsize=(7, 4))
ax.bar(regions, sales, color='#0e7490')
ax.set_title('Total sales by region')
ax.set_ylabel('Sales (Rs)')
ax.bar_label(ax.containers[0])     # value labels on bars
plt.tight_layout()
plt.show()

2Seaborn: beautiful statistical plots

Seaborn wraps Matplotlib with gorgeous defaults and one-line statistical charts. It works directly with DataFrames and understands categories.

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

# grouped bars: sales by region, split by quarter — one line
sns.barplot(data=df, x='region', y='sales', hue='quarter', errorbar=None)
plt.title('Quarterly sales by region')
plt.show()

The Seaborn staples

3Plotly: interactive charts

Static charts are great for reports; Plotly makes charts you can hover, zoom and filter — perfect for dashboards and exploration. plotly.express builds them in one line.

import plotly.express as px

fig = px.line(monthly, x='month', y='sales', markers=True,
              title='Monthly sales trend')
fig.update_traces(line_color='#0e7490')
fig.show()      # opens an interactive chart: hover for values, drag to zoom

4Choosing the right chart

The most common visualisation mistake is the wrong chart. Start from your message, then pick the chart that shows it most directly.

5Dashboard design & visual storytelling

A dashboard is not a pile of charts — it is an argument. Good design guides the eye from the headline number to the supporting detail.

Principles that make charts professional

• One message per chart. If you cannot title it in a sentence, split it.
• Maximise data-ink. Remove gridlines, borders and 3-D effects that carry no information.
• Order with meaning. Sort bars by value, not alphabetically, unless order matters.
• Highlight, do not decorate. Use one accent colour for the point you want noticed; grey the rest.
• Label directly. Put values on bars and labels on lines instead of forcing a legend hunt.
• Start bars at zero. A truncated axis exaggerates differences and misleads.

Tell a story with structure

Lead with the headline KPI (one big number), then the trend (is it improving?), then the breakdown (where is it coming from?), and finish with the so-what (the recommendation).

6Interactive dashboards with Streamlit

Streamlit turns a Python script into a shareable web app — no HTML, CSS or JavaScript. It is the fastest way to put an interactive dashboard in front of stakeholders.

# app.py
import streamlit as st
import pandas as pd
import plotly.express as px

st.title('Sales Dashboard')
df = pd.read_csv('sales.csv')

# an interactive filter
region = st.selectbox('Region', ['All'] + sorted(df['region'].unique()))
if region != 'All':
    df = df[df['region'] == region]

# a headline KPI
total = df['sales'].sum()
st.metric('Total sales', '{:,.0f}'.format(total))

# a chart that reacts to the filter
by_region = df.groupby('region', as_index=False)['sales'].sum()
st.plotly_chart(px.bar(by_region, x='region', y='sales'))

pip install streamlit
streamlit run app.py

You can now view your Streamlit app in your browser.
Local URL: http://localhost:8501

Change the dropdown and the KPI and chart update instantly — a real, interactive product built from one file.