docs/API.md

API Documentation

ModelHandler

generate(prompt, **kwargs)

Generate an image from a text prompt.

Parameters:

• prompt (str): Text prompt for generation
• negative_prompt (str, optional): Negative prompt. Default: ""
• guidance_scale (float, optional): Guidance scale. Default: 7.5
• num_inference_steps (int, optional): Number of inference steps. Default: 50
• seed (int, optional): Random seed. Default: None (random)
• width (int, optional): Image width. Default: 1024
• height (int, optional): Image height. Default: 1024
• cfg_scale (float, optional): CFG scale. Default: None (uses guidance_scale)
• lora_strength (float, optional): LoRA strength. Default: 1.0

Returns:

• tuple: (image, metadata_dict)

Example:

from src.model_handler import ModelHandler

handler = ModelHandler()
image, metadata = handler.generate(
    prompt="seamless wood texture",
    guidance_scale=7.5,
    num_inference_steps=50,
    seed=12345,
)

generate_batch(prompts, base_params=None)

Generate multiple images in batch.

Parameters:

• prompts (List[str]): List of prompts to generate
• base_params (dict, optional): Base parameters for all generations

Returns:

• Generator: Yields (image, metadata, index) tuples

Example:

prompts = ["wood texture", "fabric texture", "metal texture"]
for image, metadata, idx in handler.generate_batch(prompts):
    # Process each image
    pass

Image Processor

save_image(image, prompt, params, metadata=None)

Save an image with metadata.

Parameters:

• image (PIL.Image): Image to save
• prompt (str): Prompt used
• params (dict): Generation parameters
• metadata (dict, optional): Additional metadata

Returns:

• Path: Path to saved image file

create_thumbnail(image, size=(256, 256))

Create a thumbnail from an image.

Parameters:

• image (PIL.Image): Source image
• size (tuple): Thumbnail size (width, height)

Returns:

• PIL.Image: Thumbnail image

create_zip(files, output_path)

Create a ZIP archive from multiple files.

Parameters:

• files (List[Path]): List of file paths
• output_path (Path): Output ZIP path

Returns:

• Path: Path to created ZIP file

Presets

get_preset(name)

Get a preset by name.

Parameters:

• name (str): Preset name

Returns:

• dict or None: Preset dictionary

list_presets()

List all available preset names.

Returns:

• List[str]: List of preset names

Utils

validate_prompt(prompt, max_length=500)

Validate a prompt string.

Parameters:

• prompt (str): Prompt to validate
• max_length (int): Maximum allowed length

Returns:

• tuple: (is_valid, error_message)

validate_params(params)

Validate generation parameters.

Parameters:

• params (dict): Parameters to validate

Returns:

• tuple: (is_valid, error_message)