ConditioningType
Enums · AiDotNet.Interfaces
Types of conditioning supported by diffusion models.
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”
Fields
| Field | Summary |
|---|---|
ConditioningType Audio |
Audio conditioning (e.g., audio spectrograms). |
ConditioningType Class |
Class label conditioning (e.g., ImageNet class embeddings). |
ConditioningType Control |
Spatial control conditioning (e.g., ControlNet, T2I-Adapter). |
ConditioningType Image |
Image conditioning (e.g., CLIP vision encoder, IP-Adapter). |
ConditioningType MultiModal |
Multi-modal conditioning (combines multiple types). |
ConditioningType Text |
Text conditioning (e.g., CLIP, T5 text encoders). |
int value__ |