prospectr 0.2.8 (galo)read_nircal(): The function was still crashingbecause of some special characters in the IDs. This has been fixed (#65).
gap_der(): It can now acceptm = 0.
prospectr 0.2.7 (cakes)read_nircal(): The function was crashing because ofsome special characters in the IDs. This has been fixed (#65).
kenStone(): An issue has been fixed for thek argument whengroup was used. The functionprevented to pass a value to thek argument larger than thenumber of groups passed to thegroup argument (#51).Thanks to Michael Simmler from Agroscope for pointing at thisissue.
duplex(): Avoid error when the number of samples toselect is exactly half of the size of the input data set (#48).Thanks to Philipp Baumann for his pull request solving this issue. Somemodifications to his original solution were made which also address theconcerns raised by Matthew Dirks (skylogic004) (#56).Thank you!
read_nircal(): fixes a bug when reading some specialtype of comments of each sample in the file.
prospectr 0.2.6read_nircal(): non-utf characters in sample IDs areproperly handled (before some IDs were wrongly converted intoNAs).Non-UTF8 characters are now converted to their correspondingcodes.
Documentation is now compatible with HTML5.
prospectr 0.2.5kenStone(): an error in a sanity check thatprevented the function to run when the number of groups to be selectedwas smaller than the total number of groups. This sanity check has beenfixed. Thanks to Pogs Manalili for reporting this (#41).
kenStone(): this function was failing when whenusing the Mahalanobis distance on a matrix of one column. By default thefunction transforms the matrix onto the Mahalanobis space using eitherSVD or eigendecomposition. This makes sense for matrices with more thantwo columns. However for a matrix of one column, we now assume that suchspace is equivalent to the variable divided by the its standarddeviation. Thanks to Sergio Roldan (https://github.com/sdroldan) forreporting this (#40).
spliceCorrection(): correcting more than one pointat once (argument splice) in the spectra was returning an error. Thiswas a bug introduced in version 0.2.4. Thanks to Jose Lucas Safanelli(https://github.com/zecojls) for reporting this (#39).
prospectr 0.2.4baseline(): in some cases the function did notproperly capture the baseline confounding maximum values of a peaks aspart of the baseline. This has been addressed in this version byensuring the envelope used in the computation of the convex hull (usedto extract the baseline) is properly defined. At the edges, thisenvelope has always values higher than any peak of thespectrum.
read_nircal(): the function was wrongly identifyingactual .nir files as invalid. This preventing the function from readingany nir file. This has been fixed.
prospectr 0.2.3ThekenStone() function now includes a new featurethat allows to initialize the calibration sample search with auser-defined subset of observations (which are to be included in thefinal calibration subset). Thanks to Thorsten Behrens and PhilippBaumann for suggesting this feature.
The gap-segment derivatives (gapDer()) function nowallows for high order derivatives. In previous versions the functiononly allowed up to derivatives of 4th order, in this version the thefunction accepts as derivative order any integer larger than 1.
binning()a bug in the creation of the binning groupshas been fixed. This bug is in fact inherited from a problem in thefindInterval() function. The breaks (given in the vecarument) might get corrupted when they contain many decimal places.These breaks (in vec) are used to define the final bins. The problem inthe binning function was that when a frequency variable(e.g. wavelength) was exactly on the left of the bin range the variablewas assigned to the next bin. In some cases this lead to smalldiscrepancies in the in the computation of the mean of thebins.
spliceCorrection() now accepts one or two values asinput for the splice argument. Previously it only accepted a vector oflength two. For example, now it corrrects for splice steps of spectrathat originates from spectrometers with two detectors (i.e. it correctsfor the potential abrupt transition betwteen the twodetectors).
An extra sanity check has been added to theread_nircal() function. The function evaluaes whether it isindeed a file properly produced by the BUCHI nircal software.
There was a bug in the filter for the 3rd order derivative in gapsegment derivative (gapDer()) function. One of the factorsin the filter had a wrong negative sign.
There was a bug in theread_nircal() function. Itemerged only in some files and prevented the function from reading theDescription field properly
msc() function the argumentreference_spc has been renamed toref_spectrumto emphasize that its input must be is a vector and not a matrix ofspectra.prospectr 0.2.2baseline()The package was stripping some symbols for Rcpp functions inMakevars in order to reduce the installation size of the package. Nowthese lines have been commented to comply with CRAN policies
standardNormalVariate() returns now a matrix(previously a data.frame)
prospectr 0.2.1resample() also has a new argument(...) to pass additional arguments to thesplinefun() function (of stats) used withinresample() when the argumentinterpol = 'spline' (the default since version 0.2.0)read_nircal(), the description file was notbeing properly readprospectr 0.2.0read_nircal()) for reading buchi NIRcalfiles has been addedprospectr 0.1.4kenStone() andduplex() whenthe group parameter is usedprospectr 0.1.3prospectr 0.1.2readASD() when ASD file version = 7.0spliceCorrection()prospectr 0.1.1shenkWest() whenrm.outlier = TRUEgapDer() when input is a data.frameprospectr 0.1