Scientific computing in python is well-established. This package takes advantage of new work at Rstudio that fosters python-R interoperability. Identifying good practices of interface design will require extensive discussion and experimentation, and this package takes an initial step in this direction.
A key motivation is experimenting with an incremental PCA implementation with very large out-of-memory data. We have also provided an interface to the sklearn.cluster.KMeans procedure.
The package includes a list of references to python modules.
We can acquire python documentation of included modules with
py_help: The following result could
skd = reticulate::import("sklearn")$decomposition py_help(skd) Help on package sklearn.decomposition in sklearn: NAME sklearn.decomposition FILE /Users/stvjc/anaconda2/lib/python2.7/site-packages/sklearn/decomposition/__init__.py DESCRIPTION The :mod:`sklearn.decomposition` module includes matrix decomposition algorithms, including among others PCA, NMF or ICA. Most of the algorithms of this module can be regarded as dimensionality reduction techniques. PACKAGE CONTENTS _online_lda base cdnmf_fast dict_learning factor_analysis fastica_ incremental_pca ...
The reticulate package is designed to limit the amount of effort required to convert data from R to python for natural use in each language.
To examine a submatrix, we use the take method from numpy. The bracket format seen below notifies us that we are not looking at data native to R.
We’ll use R’s prcomp as a first test to demonstrate performance of the sklearn modules with the iris data.
We have a python representation of the iris data. We compute the PCA as follows:
## SkDecomp instance, method: PCA ## retrieve transformed data with getTransformed(), ## python reference with pyobj(), only for use with basiliskRun()
This returns an object that can be reused through python methods.
The numerical transformation is accessed via
##  150 4
## [,1] [,2] [,3] [,4] ## [1,] -2.684126 0.3193972 -0.02791483 -0.002262437 ## [2,] -2.714142 -0.1770012 -0.21046427 -0.099026550 ## [3,] -2.888991 -0.1449494 0.01790026 -0.019968390 ## [4,] -2.745343 -0.3182990 0.03155937 0.075575817 ## [5,] -2.728717 0.3267545 0.09007924 0.061258593 ## [6,] -2.280860 0.7413304 0.16867766 0.024200858
The native methods can be applied to the
Concordance with the R computation can be checked:
## PC1 PC2 PC3 PC4 ## [1,] 1 0 0 0 ## [2,] 0 -1 0 0 ## [3,] 0 0 -1 0 ## [4,] 0 0 0 -1
A computation supporting a priori bounding of memory consumption is available. In this procedure one can also select the number of principal components to compute.
This procedure can be used when data are provided in chunks, perhaps from a stream. We iteratively update the object, for which there is no container at present. Again the number of components computed can be specified.
We have extracted methylation data for the Yoruban
subcohort of CEPH from the yriMulti package. Data
from chr6 and chr17 are available in an HDF5 matrix
in this BiocSklearn package. A reference to the
dataset through the h5py File interface is created by
## <HDF5 dataset "assay001": shape (64, 44560), type "<f8">
We will explicitly define the numpy matrix.
## [] ##  64 ## ## [] ##  44560
We’ll treat genes as records and individuals as features.
We’ll define three chunks of the data and update
the partial PCA contributions in the object
Verify against the standard PCA, checking correlation between the projections to the first four PCs.
We need more applications and profiling.