dataPainter 0.1.0

Interactive 2D dataset sketching tool

DataPainter

DataPainter is an interactive tool for creating labeled 2D datasets by drawing directly on a grid. It is designed for quickly generating synthetic data for machine learning, visualization, and experimentation.

Features

• Interactive drawing of data points using mouse
• Multiple labeled classes with color coding
• Adjustable brush size and point density
• Real-time visualization on a 2D grid
• Export dataset to CSV format
• Supports up to 9 distinct labels
• Lightweight and easy to use

Installation

pip install dataPainter

---------------------------------------------
from dataPainter import create2Ddata as cd2

cd2.create(
    x="x",
    y="y",
    x_range=[0, 100],
    y_range=[0, 100],
    labels=["ClassA", "ClassB", "ClassC"],
    filename="data.csv"
)
---------------------------------------------


Controls
Right Mouse Button: Draw points
Mouse Move: Continue drawing
Left Click: Select label
[ / ]: Decrease / Increase brush size
/ =: Decrease / Increase points per stroke
1–9: Switch labels
U: Undo last stroke
S: Save dataset to CSV
Q: Quit application
Output

The generated CSV file contains:

x coordinate
y coordinate
label

Example:

x,y,label
12.5,34.2,ClassA
45.1,78.9,ClassB
Requirements
Python 3.7+
OpenCV
NumPy
Use Cases
Creating synthetic datasets for machine learning
Teaching classification concepts
Rapid prototyping of models
Data visualization experiments
Limitations
Maximum of 9 labels (mapped to keys 1–9)
Desktop-only (OpenCV-based UI)
Future Improvements
GUI-based controls (dropdowns, sliders)
Web-based interface
Export formats beyond CSV
Dataset statistics panel
License

This project is licensed under the MIT License.