The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.

EFAfactors 1.2.4

• Added - Added the method for determining the number of factos by MAP method.
• Added - Added the method for determining the number of factos by STOC method.
• Added - Added the method for determining the number of factos by LSTM, implemented through NN.
• Added - Supplemented the training example dataset for LSTM data.datasets.LSTM and the normalizing scale data.scaler.LSTM.
• Added - Added load.scaler and load.NN to extract data and models from different neural networks, and removed load_scaler and load_DNN.
• Changed - Changed the example dataset for DNN data.datasets and the normalizing scale data.scaler to data.datasets.DNN and data.scaler.DNN.
• Changed - Completely transformed the original DNN_predictor into NN, enabling it to use different neural network models for factor retention.
• Changed - Completely transformed the original extractor.feature.FNN into extractor.feature.NN, enabling it to extract features for different neural network models.
• Changed - Improved the naming conventions of functions and variables, using . globally instead of the underscore _.
• Changed – The original RMSE calculation for CD used summation. Although this did not affect the significance test, it has been updated to use the mean to align perfectly with the theoretical formulation.
• Changed - Reorganized the documentation comments for print and plot.

EFAfactors 1.2.3

• Note - Due to the correction of factor.analysis in version 1.2.3, methods involving factor.analysis may produce different outputs compared to versions prior to 1.2.3, such as CD and FF. Please take the results from the new version as the standard.
• Fixed - Fixed the issue where factor.analysis only returned nfact eigenvalues.
• Change - At the request of CRAN, some Examples have been removed to avoid excessive time consumption during automatic checks.

EFAfactors 1.2.2

• Fixed - Fixed the issue where the loadings in factor.analysis may be inverted.

EFAfactors 1.2.1

• Change - Optimized the CDF function code to improve its execution efficiency.
• Fixed - Resolved the loop error in GenData when the number of questions is very large.
• Fixed - Fixed src/Makevars to achieve better portability.

EFAfactors 1.2.0

• Fixed - The function, originally dependent on the R package ParamHelpers, has been rewritten based on version 1.14.1 to support the FF function and no longer depends on the R package ParamHelpers.
• Fixed - The function EFAsim.data no longer requires the parameter seed.
• Added - A new function check_python_libraries has been provided to help users check if the Python libraries numpy and onnxruntime are missing, and users can easily install them using this function when they are not present.

EFAfactors 1.1.1

• Fixed - Corrected the error in Hull.
• Added - Added the URL of the online manual to the DESCRIPTION field.

EFAfactors 1.1.0

• Added - Added factor analysis eigenvalues for EFAscreet, Hull, and KGC.
• Fixed - Corrected the erroneous DESCRIPTION.
• Fixed - Corrected the error in the Hull plot.
• Fixed - Corrected the calculation method for reference eigenvalues in factor analysis based PA.

EFAfactors 1.0.0

• Initial release.

These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.