Core Functions & Parameters

The Udility library provides a streamlined interface for generating high-quality, labeled educational illustrations. The primary entry point for users is the diffuser module.

Main Generation Interface

`generate_image_from_text`

The primary function for creating and displaying an illustration in a single call. It orchestrates the process of generating instructions, creating SVG code, converting it to a PNG file, and rendering it to the screen.

Parameters:

text_description (str): A detailed or simple prompt describing the educational concept or diagram you wish to generate (e.g., "The water cycle").
output_filename (str, optional): The path and filename where the resulting PNG image will be saved. Defaults to 'output.png'.

Example:

from Udility import diffuser

# Generate and display a diagram of a plant cell
diffuser.generate_image_from_text(
    text_description="Diagram of a plant cell with labels for nucleus, cell wall, and chloroplasts",
    output_filename="plant_cell.png"
)

Utility Functions

These functions are used by the main interface but can be called independently for more granular control over the image generation and display process.

`svg_to_png`

Converts raw SVG string data into a PNG file. This is useful if you have custom SVG code you wish to render using the Udility backend.

Parameters:

svg_code (str): The raw string containing valid <svg>...</svg> markup.
output_filename (str, optional): The destination path for the converted PNG file. Defaults to 'output.png'.

Example:

svg_data = '<svg height="100" width="100"><circle cx="50" cy="50" r="40" fill="red" /></svg>'
diffuser.svg_to_png(svg_data, output_filename="circle.png")

`display_image`

A helper function to render a generated PNG file using matplotlib. This is particularly useful in Jupyter Notebooks or Google Colab environments.

Parameters:

image_path (str): The file path to the PNG image you want to display.

Example:

diffuser.display_image("output.png")

Internal Workflow Functions

The following functions are used internally by generate_image_from_text. While they are accessible, they are primarily responsible for the multi-stage LLM reasoning process.

Note: These internal functions require a valid OPENROUTER_API_KEY to be set in your environment variables to communicate with the Llama 3.1 405B model.