Interface UNetModelConfig

a collection of downsample rates at which attention is applied. For example, if this contains 4, then at 4x downsampling, attention is applied.

channel multiples per down/up sampling stage

if true use learnable convolutional upsampling/downsampling

determines whether to use 1D, 2D, or 3D convolutions

the dropout probability

channels of input tensor

base channels in model

if specified, then this model will be class-conditioned with numClasses classes

if specified, ignore numHeads and instead use this number of channels in each attention head

the number of attention heads in each attention layer

number of residual blocks per down/up sampling stage

channels of output tensor

use residual blocks for up/down sampling

use a FiLM-like conditioning mechanism