Package: h2otools 0.4
E. F. Haghish
h2otools: Machine Learning Model Evaluation for 'h2o' Package
In the process of model selection, the common practice is to select a model with the higher performance. However, the fine-tuning process might tune multiple models with negligible performance differences. This software provides a statistical procedure for comparing the performance of machine learning models using a bootstrapping technique to assess significant differences between models' performances. Additionally, it offers extra performance metrics such as the F-Measure and additional functionalities for working with the H2O AI software package. For more information about the 'h2o' package, visit https://h2o.ai/.
Authors:
h2otools_0.4.tar.gz
h2otools_0.4.zip(r-4.5)h2otools_0.4.zip(r-4.4)h2otools_0.4.zip(r-4.3)
h2otools_0.4.tgz(r-4.4-any)h2otools_0.4.tgz(r-4.3-any)
h2otools_0.4.tar.gz(r-4.5-noble)h2otools_0.4.tar.gz(r-4.4-noble)
h2otools_0.4.tgz(r-4.4-emscripten)h2otools_0.4.tgz(r-4.3-emscripten)
h2otools.pdf |h2otools.html✨
h2otools/json (API)
# Install 'h2otools' in R: |
install.packages('h2otools', repos = c('https://haghish.r-universe.dev', 'https://cloud.r-project.org')) |
Bug tracker:https://github.com/haghish/h2otools/issues
Last updated 5 months agofrom:bf3ffc86b4. Checks:OK: 7. Indexed: yes.
Target | Result | Date |
---|---|---|
Doc / Vignettes | OK | Nov 22 2024 |
R-4.5-win | OK | Nov 22 2024 |
R-4.5-linux | OK | Nov 22 2024 |
R-4.4-win | OK | Nov 22 2024 |
R-4.4-mac | OK | Nov 22 2024 |
R-4.3-win | OK | Nov 22 2024 |
R-4.3-mac | OK | Nov 22 2024 |
Exports:automlModelParambootImportancebootPerformancecheckFrameFmeasuregetPerfMatrixh2o.get_idskappaperformance
Readme and manuals
Help Manual
Help page | Topics |
---|---|
AutoML Models' Parameters Summary | automlModelParam |
Bootstrap Variable Importance And Averaged Grid Variable Importance | bootImportance |
bootPerformance | bootPerformance |
check input data.frame | checkFrame |
F-Measure | Fmeasure |
getPerfMatrix | getPerfMatrix |
h2o.get_ids | h2o.get_ids |
kappa | kappa |
provides performance measures using objects from h2o | performance |