Package: layaR 0.1.0

layaR: Local Laya Decision Models via Rust

Run Laya System 1 decision models (typed choice, score and yes/no questions with calibrated probabilities) locally from R, through the laya-onnx Rust crate and ONNX Runtime. No Python or server is required at prediction time.

Authors:G. Nono Ghislain [aut, cre], Convai Innovations [cph], Microsoft Corporation [cph], Authors of the dependency Rust crates [cph]

layaR_0.1.0.tar.gz
layaR_0.1.0.tgz(r-4.6-arm64)layaR_0.1.0.tgz(r-4.5-arm64)
layaR_0.1.0.tar.gz(r-4.7-arm64)layaR_0.1.0.tar.gz(r-4.7-x86_64)layaR_0.1.0.tar.gz(r-4.6-arm64)layaR_0.1.0.tar.gz(r-4.6-x86_64)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
layaR/json (API)

# Install 'layaR' in R:
install.packages('layaR', repos = c('https://gueyenono.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/gueyenono/layar/issues

Pkgdown/docs site:https://gueyenono.github.io

On CRAN:

Conda:

rustcargo

3.30 score 30 exports 3 dependencies

Last updated from:2286bc6893 (on v0.1.0). Checks:7 OK, 8 FAIL. Indexed: yes.

TargetResultTimeFilesSyslog
linux-devel-arm64OK206
linux-devel-x86_64OK251
source / vignettesOK451
linux-release-arm64OK200
linux-release-x86_64OK238
macos-release-arm64OK282
macos-release-x86_64FAIL406
macos-oldrel-arm64OK379
macos-oldrel-x86_64FAIL409
windows-devel-arm64FAIL383
windows-devel-x86_64FAIL131
windows-release-arm64FAIL377
windows-release-x86_64FAIL150
windows-oldrel-x86_64FAIL139
wasm-releaseFAIL157

Exports:as_laya_questionslaya_annotatelaya_answerlaya_bucket_keyslaya_choicelaya_clean_emaillaya_default_enginelaya_detect_languagelaya_download_modellaya_email_statelaya_enginelaya_engine_closelaya_engine_infolaya_import_modellaya_model_availablelaya_model_pathlaya_model_rootlaya_modelslaya_noullaya_predictlaya_predict_batchlaya_predict_longlaya_presetlaya_preset_fieldslaya_probabilitieslaya_questionslaya_questions_to_jsonlaya_remove_modellaya_scorelaya_validate_questions

Dependencies:clicurljsonlite

Data frames, batches and long documents
Annotating a data frame | Abstaining on uncertain rows | Batches | Long documents | Emails | Languages

Last update: 2026-11-17
Started: 2026-11-17

Getting started with layaR
Getting a model | Asking questions | Reading the answers | When the model isn't sure | Presets | Next steps

Last update: 2026-11-17
Started: 2026-11-17

Models and performance
Where models live | Engines and memory | Threads | Parallel work | Speed | The R layer adds almost nothing | Batches: padding costs more than it saves on a CPU | Asking several questions at once | Summary

Last update: 2026-11-17
Started: 2026-11-17

Writing questions
The three question types | Choice: one label from a set | Score: a position on an ordinal scale | Noul: yes or no | Question sets | Writing questions that work | The token budget | Sharing questions with Python

Last update: 2026-11-17
Started: 2026-11-17

Readme and manuals

Help Manual

Help pageTopics
Convert to a question setas_laya_questions
Annotate a data framelaya_annotate
Bucket names for 'min_confidence'laya_bucket_keys
Clean email bodieslaya_clean_email
Use the default enginelaya_default_engine
Detect the language of stateslaya_detect_language
Download a modellaya_download_model
Build an email statelaya_email_state
Load a Laya enginelaya_engine
Inspect or close an enginelaya_engine_close laya_engine_info
Import an exported modellaya_import_model
Where models livelaya_model_available laya_model_path laya_model_root
List checkpoints and whether they are installedlaya_models
Answer typed questions about a statelaya_predict laya_predict_batch laya_predict_long predict.laya_engine
Built-in question presetslaya_preset laya_preset_fields
Build Laya questionslaya_choice laya_noul laya_question laya_score
Combine questions into a question setlaya_questions
Write a question set as JSONlaya_questions_to_json
Remove an installed modellaya_remove_model
Laya resultsas.data.frame.laya_batch_result as.data.frame.laya_result laya_answer laya_probabilities laya_result
Validate a question set without a modellaya_validate_questions