Package smile.vision.layer

package smile.vision.layer

Neural network layers for computer vision tasks.

Related Packages

Package	Description
smile.vision	Computer vision models.
smile.vision.transform	Image transformations.

Class	Description
Conv2dNormActivation	Convolution2d-Normalization-Activation block.
Conv2dNormActivation.Options	Conv2dNormActivation configurations.
FusedMBConv	Fused-MBConv replaces the depthwise-conv3×3 and expansion-conv1×1 in MBConv with single regular conv3×3.
MBConv	Mobile inverted bottleneck convolution.
MBConvConfig	EfficientNet block configuration.
SqueezeExcitation	Squeeze-and-Excitation block from "Squeeze-and-Excitation Networks".
StochasticDepth	Stochastic Depth for randomly dropping residual branches of residual architectures, from "Deep Networks with Stochastic Depth".