Optional
attentionOptional
channelchannel multiples per down/up sampling stage
Optional
contextOptional
convif true
use learnable convolutional upsampling/downsampling
Optional
dimsdetermines whether to use 1D, 2D, or 3D convolutions
Optional
dropoutthe dropout probability
Optional
dtypechannels of input tensor
base channels in model
Optional
numif specified, then this model will be class-conditioned with numClasses
classes
Optional
numif specified, ignore numHeads and instead use this number of channels in each attention head
Optional
numthe number of attention heads in each attention layer
Optional
numOptional
numnumber of residual blocks per down/up sampling stage
channels of output tensor
Optional
resblockuse residual blocks for up/down sampling
Optional
transformerOptional
useOptional
useOptional
useuse a FiLM-like conditioning mechanism
Optional
useGenerated using TypeDoc
a collection of downsample rates at which attention is applied. For example, if this contains 4, then at 4x downsampling, attention is applied.