torch_frame.nn

torch_frame.nn.encoder

FeatureEncoder	Base class for feature encoder that transforms input torch_frame.TensorFrame into (x, col_names), where x is the colum-wise PyTorch tensor of shape [batch_size, num_cols, channels] and col_names is the names of the columns.
StypeWiseFeatureEncoder	Feature encoder that transforms each stype tensor into embeddings and performs the final concatenation.
StypeEncoder	Base class for stype encoder.
EmbeddingEncoder	An embedding look-up based encoder for categorical features.
MultiCategoricalEmbeddingEncoder	An embedding look-up based encoder for multi_categorical features.
LinearEncoder	A linear function based encoder for numerical features.
LinearBucketEncoder	A numerical converter that transforms a tensor into a piecewise linear representation, followed by a linear transformation.
LinearPeriodicEncoder	A periodic encoder that utilizes sinusoidal functions to transform the input tensor into a 3-dimensional tensor.
ExcelFormerEncoder	An attention based encoder that transforms input numerical features to a 3-dimensional tensor.
LinearEmbeddingEncoder	Linear function based encoder for pre-computed embedding features.
LinearModelEncoder	Linear function based encoder with a specified model outputs embedding features.
StackEncoder	Simply stack input numerical features of shape [batch_size, num_cols] into [batch_size, num_cols, out_channels].
TimestampEncoder	TimestampEncoder for timestamp stype.

torch_frame.nn.encoding

CyclicEncoding	Cyclic encoding for input data containing values between 0 and 1.
PositionalEncoding	Positional encoding introduced in "Attention Is All You Need" paper.

torch_frame.nn.conv

TableConv	Base class for table convolution that transforms the input column-wise pytorch tensor.
FTTransformerConvs	The FT-Transformer backbone in the "Revisiting Deep Learning Models for Tabular Data" paper.
TromptConv	The Trompt cell introduced in the "Trompt: Towards a Better Deep Neural Network for Tabular Data" paper.
ExcelFormerConv	The ExcelFormer Layer introduced in the "ExcelFormer: A Neural Network Surpassing GBDTs on Tabular Data" paper.
TabTransformerConv	The TabTransformer Layer introduced in the "TabTransformer: Tabular Data Modeling Using Contextual Embeddings" paper.

torch_frame.nn.decoder

Decoder	Base class for decoder that transforms the input column-wise PyTorch tensor into output tensor on which prediction head is applied.
TromptDecoder	The Trompt downstream introduced in "Trompt: Towards a Better Deep Neural Network for Tabular Data" paper.
ExcelFormerDecoder	The ExcelFormer decoder introduced in the "ExcelFormer: A Neural Network Surpassing GBDTs on Tabular Data" paper.

torch_frame.nn.models

Trompt	The Trompt model introduced in the "Trompt: Towards a Better Deep Neural Network for Tabular Data" paper.
FTTransformer	The FT-Transformer model introduced in the "Revisiting Deep Learning Models for Tabular Data" paper.
ExcelFormer	The ExcelFormer model introduced in the "ExcelFormer: A Neural Network Surpassing GBDTs on Tabular Data" paper.
TabNet	The TabNet model introduced in the "TabNet: Attentive Interpretable Tabular Learning" paper.
ResNet	The ResNet model introduced in the "Revisiting Deep Learning Models for Tabular Data" paper.
TabTransformer	The Tab-Transformer model introduced in the "TabTransformer: Tabular Data Modeling Using Contextual Embeddings" paper.
MLP	The light-weight MLP model that mean-pools column embeddings and applies MLP over it.