Hi,
I am using Halla version 0.8.17 and noticed some inconsistencies between this version and the documentation at https://github.com/biobakery/halla#markdown-header-basic-usage
e.g. in the docs, there is the following
usage: halla [-h] [--version] -X <input_dataset_1.txt>
[-Y <input_dataset_2.txt>] -o <output> [-q <.1>]
[-p {ecdf,gpd,none}] [-a {HAllA,AllA}] [-i <1000>]
[-m {nmi,ami,mic,dmic,dcor,pearson,spearman}]
[-d {none,mca,pca,ica,cca,kpca,pls,medoid}]
[--fdr {bh,by,bonferroni,no_adjusting}] [-v VERBOSE]
[--diagnostics-plot] [--discretizing {equal-area,hclust,none}]
[--linkage {single,average,complete,weighted}]
[--apply-stop-condition] [--generate-one-null-samples] [--header]
[--format-feature-names] [--nproc <1>] [--nbin <None>] [-s SEED]
[-e ENTROPY_THRESHOLD] [-e1 ENTROPY_THRESHOLD1]
[-e2 ENTROPY_THRESHOLD2] [--missing-char MISSING_CHAR]
[--missing-method {mean,median,most_frequent}]
[--missing-data-category] [--write-hypothesis-tree]
[-t {log,sqrt,arcsin,arcsinh,}]
and in the version I am running, I see:
usage: halla [-h] [--version] -X <input_dataset_1.txt>
[-Y <input_dataset_2.txt>] -o <output> [-q <.1>] [--fnt <.25>]
[-p {ecdf,gpd,none}] [-a {HAllA,AllA}] [-i <1000>]
[-m {nmi,ami,mic,dmic,dcor,pearson,spearman,r2,chi,mi}]
[--fdr {bh,by,y,meinshausen,bonferroni,no_adjusting}]
[-v VERBOSE] [--diagnostics-plot]
[--discretizing {equal-freq,hclust,jenks,none}]
[--linkage {single,average,complete,weighted}]
[--generate-one-null-samples] [--header] [--format-feature-names]
[--nproc <1>] [--nbin <None>] [-s SEED] [-e ENTROPY_THRESHOLD]
[-e1 ENTROPY_THRESHOLD1] [-e2 ENTROPY_THRESHOLD2]
[--missing-char MISSING_CHAR]
[--fill-missing {mean,median,most_frequent}]
[--missing-data-category] [--write-hypothesis-tree]
[-t {log,sqrt,arcsin,arcsinh,}]
In particular, I would like to know what happened to the decomposition option? If this has been removed, what decomposition method is halla now using as standard?
Thanks,
Ramiro