nn_tokenizer_num: Numeric Tokenizer

In mlr-org/mlr3torch: Deep Learning with 'mlr3'

Numeric Tokenizer

Description

Tokenizes numeric features into a dense embedding. For an input of shape ⁠(batch, n_features)⁠ the output shape is ⁠(batch, n_features, d_token)⁠.

Usage

nn_tokenizer_num(n_features, d_token, bias, initialization)

Arguments

n_features	(integer(1)) The number of features.
d_token	(integer(1)) The dimension of the embedding.
bias	(logical(1)) Whether to use a bias.
initialization	(character(1)) The initialization method for the embedding weights. Possible values are "uniform" and "normal".

References

Gorishniy Y, Rubachev I, Khrulkov V, Babenko A (2021). “Revisiting Deep Learning for Tabular Data.” arXiv, 2106.11959.

mlr-org/mlr3torch documentation built on April 17, 2025, 8:22 p.m.