ResNet

NeuroTabModels.Models.ResNet.ResNetConfig Type

ResNetConfig(; kwargs...)

Configuration for a residual MLP backbone for tabular data.

Arguments

• stack_size::Int: Number of residual blocks (default 1).

• hidden_size::Int: Hidden dimension (default 64).

• act::Symbol: Activation — :relu, :gelu, :sigmoid, or :tanh (default :relu).

• dropout::Float64: Dropout rate within each block (default 0.0).

• MLE_tree_split::Bool: Split output head for Gaussian MLE (default false).

source

NeuroTabModels.Models.ResNet.ResNetConfig Method

(config::ResNetConfig)(; nfeats, outsize)

Build a Lux.Chain from config.

Each block applies two Dense layers with batch norm and a skip connection.

Arguments

• nfeats::Int: Number of input features.

• outsize::Int: Number of output units.

Returns

A Lux.Chain with residual blocks.

source