What is new
31.03.2023
- Added section about normalization to Preprocessing chapter.
- Added section about missing values replacement to Preprocessing chapter.
- Improved description of validation procedure in PCA, PLS and other methods, including the use of Procrustes cross-validation.
- Small improvements in selected parts of text.
09.11.2022
Improved description of iPLS section, including new features appeared in v. 0.13.1.
03.03.2022
Added section explaining how to use
pcv.nareplace()
methodAdded separate section for Procrustes cross-validation.
14.07.2022
Added description of new features appeared in v. 0.13.0 of the package, including new plotting method plotRMSERatio()
for PLS models and new cross-validation possibilities. Validation of PLS model text has been extended and made as separate section.
12.09.2021
Added description of new features appeared in v. 0.12.0 of the package, including:
- new preprocessing methods
prep.transform()
,prep.varsel()
. - improvements to
prep.savgol()
andprep.norm()
. - a possibility to combine several preprocessing method together and apply at once.
In addition to that, text for Datasets chapter has been revised to provide more clear guides while the part about preprocessing was detached and now is available as separate chapter where you can find all details. Description of ALS baseline correction was moved to the section Correction of baseline.
23.04.2021
Added description of new features added in v. 0.11.4:
- using test set validation in iPLS.
- possibility to force selected contribution and spectral values in MCR-ALS.
- added information about how to provide user defined indices for pure variables in Purity method
- added link to relevant paper in the PCA model complexity section
22.01.2021
- added section about Procrustes cross-validation for PCA, which was recently added to the package (v. 0.11.3).
23.07.2020
- section about MCR Purity method has been re-written to correspond to the algorithm changes in v. 0.11.1.
12.07.2020
- added chapters about MCR methods implemented in v. 0.11.0.
- added description of new preprocessing methods
prep.alsbasecorr()
- several small improvements in text