BaseModels

This module provides foundational classes and architectures for deeptab models, including various neural network architectures tailored for tabular data.

Modules	Description
Mambular	Flexible neural network model leveraging the Mamba architecture with configurable normalization techniques for tabular data.
MLP	Multi-layer perceptron (MLP) model designed for tabular tasks, initialized with a custom configuration.
ResNet	Deep residual network (ResNet) model optimized for structured/tabular datasets.
FTTransformer	Feature Tokenizer (FTTransformer) model for tabular tasks, incorporating advanced embedding and normalization techniques.
TabTransformer	TabTransformer model leveraging attention mechanisms for tabular data processing.
NODE	Neural Oblivious Decision Ensembles (NODE) for tabular tasks, combining decision tree logic with deep learning.
TabM	TabM architecture designed for tabular data, implementing batch-ensembling MLP techniques.
NDTF	Neural Decision Tree Forest (NDTF) model for tabular tasks, blending decision tree concepts with neural networks.
TabulaRNN	Recurrent neural network (RNN) model, including LSTM and GRU architectures, tailored for sequential or time-series tabular data.
MambAttention	Attention-based architecture for tabular tasks, combining feature importance weighting with advanced normalization techniques.
SAINT	SAINT model. Transformer based model using row and column attention.
MambaTab	Tabular model using a Mamba-Block on a joint input representation.
AutoInt	Automatic Feature Interaction model for tabular data.
ENODE	Embedding Neural Oblivious Decision Ensembles for tabular tasks.
ModernNCA	Modern Nearest Centroid Approach for tabular deep learning.
Tangos	Tangos model for tabular data.
Trompt	Trompt model for tabular data.

• deeptab.base_models