Use R in Python

Master R-Python Integration: Learn How to Use R in Python with rpy2 Package – Install Guide, Practical Examples, Data Frame Tutorials, and Advanced Techniques for Data Scientists.

There are several ways to use R from Python. Since the post is about using R in Python, here are the most common and effective methods. First of all, we need to have both R and Python installed. Choose the method based on your needs: rpy2 for tight integration, subprocess for simple script execution, or consider if you can use Python equivalents instead. The post ‘Use R in Python’ is presented for those who already know R and want to call it from within Python to use the advanced Pandas data manipulation tools.

Install rpy2 Package

The rpy2 is the primary package for integrating R with Python. To install rpy2 in Python, use the following code:

pip install rpy2

Note that you also need R installed on your system.

What is rpy2 Package?

rpy2 is a powerful Python package that provides a seamless bidirectional interface between Python and R, allowing you to run R code, use R packages, and manipulate R objects directly from Python.

What are the Key Features of rpy2 Package?

Execute R code from Python scripts
Import R packages as Python modules
Convert data between Python and R formats
Access R objects as Python objects (and vice versa)
Memory-efficient data sharing
Interactive R console in Python

When to use rpy2 Package?

Use rpy2 in the following situations:

You need specific R packages not available in Python
Your team knows R but needs Python integration
You require advanced statistical models
You are migrating from R to Python gradually
You need publication-quality R graphics

Give a Basic Working Example that makes use of rpy2 Package

The following code performs basic computation, such as the computation of the average value of a vector, by making use of R in Python.

import rpy2.robjects as ro
from rpy2.robjects import pandas2ri
from rpy2.robjects.packages import importr

# Convert Pandas DataFrame to R dataframe
pandas2ri.activate()

# Load R packages
base = importr('base')
stats = importr('stats')

# Execute R code
ro.r('''
    x <- c(1, 2, 3, 4, 5)
    avg <- mean(x)
    print(avg)
''')

# Create R objects from Python
r_vector = ro.FloatVector([1, 2, 3, 4, 5])
mean_result = ro.r.mean(r_vector)
print(f"Mean: {mean_result[0]}")

Working with Data Frames: Use R in Python

The py2rpy() from rpy2.robjects convert the pandas dataframe to the R Language.

import pandas as pd
from rpy2.robjects import pandas2ri

# Convert pandas DataFrame to R
pandas2ri.activate()
df_python = pd.DataFrame({'x': [1, 2, 3], 'y': [4, 5, 6]})
r_df = pandas2ri.py2rpy(df_python)

# Use R functions
ro.r.assign('r_df', r_df)
summary = ro.r('summary(r_df)')
print(summary)

Using subprocess

The simple method for running an R script is

import subprocess
import json

# Run R script
result = subprocess.run(
    ['Rscript', 'my_script.R'],
    capture_output=True,
    text=True
)
print(result.stdout)

# Pass data via JSON
data = {'x': [1, 2, 3], 'y': [4, 5, 6]}
with open('input.json', 'w') as f:
    json.dump(data, f)

subprocess.run(['Rscript', 'process_data.R'])

Using R Markdown/ Jupyter Notebooks

Embed Python chunks in R Markdown or use Jupyter notebooks with both kernels.

Give an Advanced example that makes use of rpy2 Package

The following example creates random sample data from the standard normal probability distribution using the NumPy Python Library. The NumPy objects are converted to an R data frame, then a simple regression line is fitted using R syntax, and finally, a regression plot is drawn using the ggplot2 package.

import rpy2.robjects as ro
from rpy2.robjects import pandas2ri
from rpy2.robjects.packages import importr
import pandas as pd
import numpy as np

# Activate pandas conversion
pandas2ri.activate()

# Import R packages
stats = importr('stats')
ggplot2 = importr('ggplot2')

# Create sample data
np.random.seed(42)
df = pd.DataFrame({
    'x': np.random.randn(100),
    'y': np.random.randn(100) * 0.5 + 2
})

# Convert to R dataframe
r_df = pandas2ri.py2rpy(df)

# Run linear regression in R
ro.r.assign('df_r', r_df)
lm_result = ro.r('lm(y ~ x, data=df_r)')
summary = ro.r('summary(lm_result)')
print(summary)

# Create plot
ro.r('''
    p <- ggplot(df_r, aes(x=x, y=y)) +
         geom_point() +
         geom_smooth(method="lm") +
         ggtitle("Linear Regression in R from Python")
    print(p)
''')

R in Python Tutorial: How to Use rpy2 Package for Data Science Integration – Complete Guide Covering Installation, Basic to Advanced Examples, Data Frame Conversion, and R Markdown Methods. Get Instant Access to Code Samples!

Statistics and Data Analysis

Table of Contents