Package: tabnet 0.6.0.9000

Christophe Regouby

tabnet: Fit 'TabNet' Models for Classification and Regression

Implements the 'TabNet' model by Sercan O. Arik et al. (2019) <doi:10.48550/arXiv.1908.07442> with 'Coherent Hierarchical Multi-label Classification Networks' by Giunchiglia et al. <doi:10.48550/arXiv.2010.10151> and provides a consistent interface for fitting and creating predictions. It's also fully compatible with the 'tidymodels' ecosystem.

Authors:Daniel Falbel [aut], RStudio [cph], Christophe Regouby [cre, ctb], Egill Fridgeirsson [ctb], Philipp Haarmeyer [ctb], Sven Verweij [ctb]

tabnet_0.6.0.9000.tar.gz
tabnet_0.6.0.9000.zip(r-4.5)tabnet_0.6.0.9000.zip(r-4.4)tabnet_0.6.0.9000.zip(r-4.3)
tabnet_0.6.0.9000.tgz(r-4.4-any)tabnet_0.6.0.9000.tgz(r-4.3-any)
tabnet_0.6.0.9000.tar.gz(r-4.5-noble)tabnet_0.6.0.9000.tar.gz(r-4.4-noble)
tabnet_0.6.0.9000.tgz(r-4.4-emscripten)tabnet_0.6.0.9000.tgz(r-4.3-emscripten)
tabnet.pdf |tabnet.html✨
tabnet/json (API)
NEWS

# Install 'tabnet' in R:

install.packages('tabnet', repos = c('https://mlverse.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/mlverse/tabnet/issues

On CRAN:

tabnet

32 exports 109 stars 4.57 score 99 dependencies 62 scripts 1.2k downloads

Last updated 2 months agofrom:c8c82d24ab. Checks:OK: 7. Indexed: yes.

Target	Result	Date
Doc / Vignettes	OK	Aug 28 2024
R-4.5-win	OK	Aug 28 2024
R-4.5-linux	OK	Aug 28 2024
R-4.4-win	OK	Aug 28 2024
R-4.4-mac	OK	Aug 28 2024
R-4.3-win	OK	Aug 28 2024
R-4.3-mac	OK	Aug 28 2024

Exports:%>%attention_width cat_emb_dim check_compliant_node checkpoint_epochs decision_width drop_last encoder_activation feature_reusage lr_scheduler mask_type mlp_activation mlp_hidden_multiplier momentum nn_prune_head.tabnet_fit nn_prune_head.tabnet_pretrain node_to_df num_independent num_independent_decoder num_shared num_shared_decoder num_steps optimizer penalty tabnet tabnet_config tabnet_explain tabnet_fit tabnet_nn tabnet_pretrain verbose virtual_batch_size

Dependencies:bit bit64 callr class cli clock codetools colorspace coro cpp11 crayon data.table data.tree desc diagram dials DiceDesign digest doFuture dplyr fansi farver foreach furrr future future.apply generics ggplot2 globals glue gower GPfit gtable hardhat hms ipred isoband iterators jsonlite KernSmooth labeling lattice lava lhs lifecycle listenv lubridate magrittr MASS Matrix mgcv modelenv munsell nlme nnet numDeriv parallelly parsnip pillar pkgconfig prettyunits processx prodlim progress progressr ps purrr R6 RColorBrewer Rcpp recipes rlang rpart rsample safetensors scales sfd shape slider SQUAREM stringi stringr survival tibble tidyr tidyselect timechange timeDate torch tune tzdb utf8 vctrs viridisLite warp withr workflows yardstick zeallot

Fitting tabnet with tidymodels

Rendered fromtidymodels-interface.Rmdusingknitr::rmarkdownon Aug 28 2024.

Last update: 2023-02-06
Started: 2020-12-18

Hierarchical Classification

Rendered fromHierarchical_classification.Rmdusingknitr::rmarkdownon Aug 28 2024.

Last update: 2023-12-04
Started: 2023-07-22

Interpretation tools

Rendered frominterpretation.Rmdusingknitr::rmarkdownon Aug 28 2024.

Last update: 2024-02-16
Started: 2021-01-07

Self-supervised training and fine-tuning

Rendered fromselfsupervised_training.Rmdusingknitr::rmarkdownon Aug 28 2024.

Last update: 2023-07-22
Started: 2023-07-22

Training a Tabnet model from missing-values dataset

Christophe Regouby

Rendered fromMissing_data_predictors.Rmdusingknitr::rmarkdownon Aug 28 2024.

Last update: 2024-02-16
Started: 2022-02-11

Citation

Development and contributors

Readme and manuals

Help Manual

Help page	Topics
Parameters for the tabnet model	attention_width decision_width feature_reusage mask_type momentum num_independent num_shared num_steps
Plot tabnet_explain mask importance heatmap	autoplot.tabnet_explain
Plot tabnet_fit model loss along epochs	autoplot.tabnet_fit autoplot.tabnet_pretrain
Non-tunable parameters for the tabnet model	cat_emb_dim checkpoint_epochs drop_last encoder_activation lr_scheduler mlp_activation mlp_hidden_multiplier num_independent_decoder num_shared_decoder optimizer penalty verbose virtual_batch_size
Check that Node object names are compliant	check_compliant_node
Prune top layer(s) of a tabnet network	nn_prune_head.tabnet_fit nn_prune_head.tabnet_pretrain
Turn a Node object into predictor and outcome.	node_to_df
Parsnip compatible tabnet model	tabnet
Configuration for TabNet models	tabnet_config
Interpretation metrics from a TabNet model	tabnet_explain tabnet_explain.default tabnet_explain.model_fit tabnet_explain.tabnet_fit tabnet_explain.tabnet_pretrain
Tabnet model	tabnet_fit tabnet_fit.data.frame tabnet_fit.default tabnet_fit.formula tabnet_fit.Node tabnet_fit.recipe
TabNet Model Architecture	tabnet_nn
Tabnet model	tabnet_pretrain tabnet_pretrain.data.frame tabnet_pretrain.default tabnet_pretrain.formula tabnet_pretrain.Node tabnet_pretrain.recipe