oliviergarciadev / fanalysis Goto Github PK

Hi,

I don't understand how std_unit=False work, how to do the same in creating a PCA class instance with scikit?

I"m converting a R code that use decorana to Python,
and found fanalysis as alternative to decorana with (ira=1).
Results from R, for the 4 eigen values are same.
But, I did not found how to get Row and Col projections. ??

For example,
in R decorana we coud use:

library("vegan")
X <-read.table("example.tsv", header=TRUE, sep='\t')
ca_ra <- decorana(X, ira=1)
# show col projection
ca_ra$cproj
         RA1        RA2         RA3          RA4
A  1.93003452 -3.4006348 -0.41834170  5.741832353
B  1.93004285 -3.4006350 -0.41834167  5.741832392
C  0.03164434  0.4015328  3.19055179  0.051847893
D  0.60775895 -0.8324331 -0.30835257  0.040141236

where A,B,C,D is columns names.

Hi,

It looks like as_matrix() has been deprecated since pandas 0.23 and finally been removed.

Would you be open to a pull request that update the code, and if so will you publish an updated version on PyPI ?

Thanks.