rpy2-Matrix: mapping R’s package Matrix for rpy2¶

Installation¶

pip install git+https://github.com/rpy2/rpy2-Matrix.git#egg=rpy2_Matrix

Usage¶

This package is a wrapper for the R package Matrix.

Create R Matrix object with Python¶

from rpy2_Matrix import Matrix
from rpy2.robjects.vectors import IntVector

m = Matrix.sparseMatrix.new(
    IntVector([1, 3, 4]), IntVector([2, 3, 5]),
    x=IntVector([3, 9, 21]), giveCsparse=True)

m is a Python object acting as a proxy to interact with an R object in the R memory space.

Map R Matrix object creating in R¶

An extension for the rpy2 conversion system is provided, allowing to map R objects resulting from the evaluation of R code to the wrapping classes defined in this package:

import rpy2_Matrix.conversion as matrix_conversion
from rpy2 import robjects

with conversion.rs4map_context:
     # The package Matrix is already imported in the embedded R
     # (implicit when importing our Python package Matrix).
     m = robjects.r('Matrix(c(0,1,0,0), 6, 4)')

m was automatically mapped to one of the matrix classes defined in this package.

Using matrix objects¶

Objects can be passed to R functions compatible with Matrix objects. The R package Matrix contains such functions.

We create a matrix to run our examples:

m = Matrix.sparseMatrix.new(
        IntVector([1, 3, 4]), IntVector([2, 3, 5]),
        x=IntVector([3, 9, 21]), giveCsparse=True)

>>> print(m)
4 x 5 sparse Matrix of class "dgCMatrix"

[1,] . 3 . .  .
[2,] . . . .  .
[3,] . . 9 .  .
[4,] . . . . 21

To get the diagonal of our matrix:

>>> Matrix.Matrix_pack.diag(m)
R object with classes: ('numeric',) mapped to:
[0.000000, 0.000000, 9.000000, 0.000000]

Values in the matrix can also be accessed through R’s own subsetting system

Note

See the rpy2 documentation for more information it in the section about the method rx().

For example, to get the values in the first row and second column (remember: R arrays are one-indexed, not zero-indexed)

>>> m.rx(1, 2)
R object with classes: ('numeric',) mapped to:
[3.000000]

To get the complete second column:

>>>  m.rx(True, 2)
R object with classes: ('numeric',) mapped to:
[3.000000, 0.000000, 0.000000, 0.000000]

Inheritance Diagram¶

module Matrix¶

Map the R package Matrix for rpy2.

This module maps some of the classes defined by the R package Matrix.

module conversion¶

This module handles the conversion of data structures.

In the future it fill ensure the conversion between R objects from the R package Matrix and scipy sparse matrices.