Skip to main content

GraphVision AI 📊👁️

GraphVision AI is a lightweight, powerful computer vision library for automatic graph classification and structured data extraction.

Built with PyTorch and EasyOCR, it is designed to look at an image of a chart, instantly recognize what kind of graph it is, and extract its labels and values into a clean, developer-friendly JSON format.


✨ Key Features

🚀 Zero-Configuration

Models and weights are automatically downloaded from Hugging Face the first time you run it. No manual weight management required.

🧠 Intelligent Routing

Automatically classifies the input image (Pie, Vertical Bar, Horizontal Bar, Line, etc.) and routes it to the correct extraction algorithm.

🖼 Robust Input Handling

Pass a file path (String), an OpenCV image (NumPy array), or a PIL Image directly into the analyzer.

🔍 Smart OCR Masking

Uses contrast filtering and spatial mapping to accurately match text labels with their corresponding graphical data points.


📦 Installation

Install directly from PyPI:

pip install graphvision-ai

🚀 Quick Start

Extracting data from a graph takes less than 5 lines of code:

from graphvision.extractor import GraphExtractor

try:
    # 1. Initialize your engine (this will download weights if needed)
    vision_engine = GraphExtractor()
    
    # Path to your test image
    image_to_test = "hbar2.png" 
    
    # 2. Run the extraction using the new method name
    print(f"\n🚀 Extracting data from {image_to_test}...")
    result_json_string = vision_engine.extract(image_to_test)
    
    # 3. Print the result (it's already a nicely formatted string!)
    print("\n✅ Extraction Successful!")
    print(result_json_string)

except Exception as e:
    print(f"\n❌ Error during testing: {e}")

📄 Example Output

{
    "type": "pie",
    "title": "Favorite Programming Languages",
    "data": {
        "Python": 45.2,
        "JavaScript": 25.1,
        "C++": 15.4,
        "Java": 14.3
    }
}

📈 Supported Graph Types

Currently, GraphVision AI supports high-accuracy extraction for:

  • pie — Pie Charts
  • vbar_categorical — Vertical Bar Charts
  • hbar_categorical — Horizontal Bar Charts

Line and Dot-Line charts coming soon.

Metadata

Release files for graphvision-ai 0.2.5

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for graphvision-ai 0.2.5
File Size Uploaded Provenance Actions
graphvision_ai-0.2.5.tar.gz 12.8 kB Without provenance Details for graphvision_ai-0.2.5.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for graphvision-ai 0.2.5
File Interpreter ABI Platform Actions
graphvision_ai-0.2.5-py3-none-any.whl Python 3 none any Details for graphvision_ai-0.2.5-py3-none-any.whl

Total release size: 24.5 kB

Release files / graphvision_ai-0.2.5.tar.gz

Download URL graphvision_ai-0.2.5.tar.gz
Size 12.8 kB
Tags Source
SHA-256 checksum
How to use checksums
56c937f179d82207cffe6e3d09b20007c09153e188489946d19f2f9a74b8b983
BLAKE2b-256 checksum
How to use checksums
b11bc33081f97f51144c174ae9d5d452babe9560840d480ca561eb0334dcb563
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.12.12

Release files / graphvision_ai-0.2.5-py3-none-any.whl

Download URL graphvision_ai-0.2.5-py3-none-any.whl
Size 11.8 kB
Tags Python 3
SHA-256 checksum
How to use checksums
9d5a001ab509cbfbe8a393dce791abfb5e8c500bcbc20b1c67e4e931270d8dc5
BLAKE2b-256 checksum
How to use checksums
c9f84c96125da523413a23987d6c3b706284dfa8a7062e18e9dcd9765d55f8e7
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.12.12

Security for graphvision-ai 0.2.5

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page