Class DiffusionTrialConfig<T>
Configuration for a diffusion model trial in AutoML.
public class DiffusionTrialConfig<T>
Type Parameters
TThe numeric type used for calculations.
- Inheritance
-
DiffusionTrialConfig<T>
- Inherited Members
Properties
BaseChannels
Gets or sets the base channels for the noise predictor.
public int BaseChannels { get; set; }
Property Value
GuidanceScale
Gets or sets the guidance scale for classifier-free guidance.
public double GuidanceScale { get; set; }
Property Value
InferenceSteps
Gets or sets the number of inference steps.
public int InferenceSteps { get; set; }
Property Value
LatentDim
Gets or sets the latent dimension (channels).
public int LatentDim { get; set; }
Property Value
LatentHeight
Gets or sets the latent spatial height (default: 64 for 512x512 images with 8x downscaling).
public int LatentHeight { get; set; }
Property Value
LatentWidth
Gets or sets the latent spatial width (default: 64 for 512x512 images with 8x downscaling).
public int LatentWidth { get; set; }
Property Value
LearningRate
Gets or sets the learning rate for training.
public double LearningRate { get; set; }
Property Value
NoisePredictorType
Gets or sets the noise predictor type.
public NoisePredictorType NoisePredictorType { get; set; }
Property Value
NumResBlocks
Gets or sets the number of residual blocks per level.
public int NumResBlocks { get; set; }
Property Value
SchedulerType
Gets or sets the scheduler type.
public DiffusionSchedulerType SchedulerType { get; set; }
Property Value
Seed
Gets or sets the optional random seed.
public int? Seed { get; set; }
Property Value
- int?
Methods
FromDictionary(Dictionary<string, object>)
Creates a configuration from a dictionary of parameters.
public static DiffusionTrialConfig<T> FromDictionary(Dictionary<string, object> parameters)
Parameters
parametersDictionary<string, object>
Returns
ToDictionary()
Converts the configuration to a dictionary of parameters.
public Dictionary<string, object> ToDictionary()