ChartMind

Natural language to chart generator -- a Python library that generates beautiful charts from natural language descriptions of your data.

---

Architecture

graph LR
    A[Natural Language Query] --> B[QueryParser]
    B --> C[ChartConfig]
    D[CSV / JSON / DataFrame] --> E[DataLoader]
    E --> F[pandas DataFrame]
    C --> G[ChartRenderer]
    F --> G
    G --> H[Bar Chart]
    G --> I[Line Chart]
    G --> J[Pie Chart]
    G --> K[Scatter Plot]
    G --> L[Histogram]
    G --> M[Heatmap]
    H --> N[PNG / SVG Output]
    I --> N
    J --> N
    K --> N
    L --> N
    M --> N

Loading

Quickstart

Installation

pip install chartmind

Python API

from chartmind import ChartGenerator

gen = ChartGenerator()

# Generate a chart from a natural language query + CSV file
config = gen.generate(
    query="Show total revenue by month as a bar chart",
    data="sales.csv",
    output="chart.png",
)
print(config.chart_type)  # "bar"
print(config.title)       # "Sum Revenue by Month"

With a DataFrame

import pandas as pd
from chartmind import ChartGenerator

df = pd.DataFrame({
    "region": ["North", "South", "East", "West"],
    "sales": [120, 98, 150, 130],
})

gen = ChartGenerator()
config = gen.generate(
    query="pie chart of sales by region",
    data=df,
    output="pie.png",
)

Config-only (no rendering)

config = gen.generate_config(
    "average score by department",
    columns=["score", "department"],
)
# Returns a plain dict: {"chart_type": "bar", "aggregation": "mean", ...}

CLI

chartmind "show revenue trend over time by month" --data sales.csv --output trend.png

Supported Chart Types

Type	Trigger Keywords
bar	bar, compare, column chart
line	line, trend, over time, timeline
scatter	scatter, correlation, relationship, xy
pie	pie, proportion, share, breakdown
histogram	histogram, distribution, frequency
heatmap	heatmap, heat map, matrix, density

Features

• Auto-detect the best chart type from natural language queries
• Keyword-based NLP -- no ML dependencies, fast and deterministic
• Supports aggregation (sum, mean, count, min, max)
• Filter clauses ("where region equals North")
• Six chart types: bar, line, scatter, pie, histogram, heatmap
• CSV and JSON data loading via pandas
• High-quality matplotlib output (PNG, SVG, PDF)
• Pydantic-validated configuration models
• CLI powered by Typer

Development

# Clone the repository
git clone https://github.com/MukundaKatta/ChartMind.git
cd ChartMind

# Install in development mode
make install-dev

# Run tests
make test

# Run linter
make lint

# Type checking
make typecheck

Contributing

See CONTRIBUTING.md for guidelines.

License

MIT License -- 2026 Officethree Technologies

---

Inspired by data visualization AI trends

Built by Officethree Technologies | Made with ❤️ and AI