Enum ConditioningType

Namespace

AiDotNet.Interfaces

Assembly

AiDotNet.dll

Types of conditioning supported by diffusion models.

public enum ConditioningType

Fields

Audio = 2

Audio conditioning (e.g., audio spectrograms).

Class = 4

Class label conditioning (e.g., ImageNet class embeddings).

Control = 3

Spatial control conditioning (e.g., ControlNet, T2I-Adapter).

Image = 1

Image conditioning (e.g., CLIP vision encoder, IP-Adapter).

MultiModal = 5

Multi-modal conditioning (combines multiple types).

Text = 0

Text conditioning (e.g., CLIP, T5 text encoders).

Remarks

For Beginners: This describes what kind of input you're using to guide generation: - Text: Written descriptions like "a beautiful sunset" - Image: Reference images for style or content - Audio: Sound clips for audio generation - Control: Structural guidance like edges, poses, or depth maps - Class: Simple class labels like "dog" or "car"