R VS Python in Data science: The winner is ...

R VS Python in Data science: The winner is ...

In the "Best" data Science tools game, R and Python have their own pros and cons. The choice between the two depends on the use of backgrounds, learning costs and other common tools needed

Martijn Theuwissen published in Datacamp.

At Datacamp, students often ask us about their daily data analysis tasks using R or Python. While we are primarily offering interactive r tutorials, we always answer this question depending on the type of data analysis challenge they face.
Both R and Python are popular statistical programming languages. The function of R is developed by statisticians (think of R's powerful data visualization capabilities), and Python is often praised for its easy-to-understand syntax.

In this article, we focus on the differences between R and Python, and how they occupy the world of data science and statistics. If you prefer visual presentation, make sure to check out the corresponding infographic "Data Science War: R vs Python".

Introduction to R

Ross Ihaka and Robert gentleman created the open source language R as an implementation of the S programming language in 1995. The aim is to develop a language that provides better and more humane mathematical analysis, statistics and graphical models. Initially, R was primarily used in academia and research, R has also been used recently in the rest of the world. This makes R the fastest growing statistical language worldwide.

The main advantage of R is its strong community, supported by mailing lists, user contributed documents and a very active stack overflow team. There is also Cran, a huge resource library where users can easily contribute R packages. These packages are a collection of r functions and data, Immediate access to the latest technologies and features without the need to develop from scratch.

Finally, if you are an experienced programmer, learning R may not be difficult. However, as a beginner, you may feel very tangled. Fortunately, there are many current learning resources that you can consult.

Introduction to Python

Python was created by Guido van Rossum 1991, emphasizing efficiency and readability of the code. Programmers who want to drill down into data analysis or apply statistics are some of the main users of Python for statistics.

The closer you get to working in an engineering environment, the more likely you are to prefer python. It is a flexible language that focuses on readability and simplicity, and its learning curve is relatively low.

Like R, Python also has packages. PyPI is the index of the Python package, consisting of the user-contributed libraries. Just like R,python has a great community, but it's a little scattered because it's a universal language. However, data science in Python is rapidly claiming to be more dominant in the Python world: applications that are expected to be growing and more innovative scientific data will see their origins here.

R and Python: Common volume

On the web, you can find lots of comparisons between R and Python usage and popularity. While these numbers often tell us that both languages are well developed in the entire ecosystem of computer science, this is hard to compare horizontally. The main reason for this is that you will only find R in the data Science environment, and on the other hand, Python is widely used as a common language in many fields, such as web development. This tends to favor ranking results in Python, and the result is some negative impact.

When and how do I use r?

R is primarily used when data analysis tasks require individual servers to be calculated or analyzed independently. Exploratory work is great, handy for almost any type of data analysis, because a lot of packages and easy-to-use tests often provide you with the tools you need to get up and running quickly. R can even be part of a big data solution.

When starting to use R, a good first step is to install the magical IDE RStudio. Once this is done, we recommend that you look at the following popular packages:

dplyr，plyr和data.table轻松地操纵包,  stringr操作字符串,zoo处理定期和不定期的时间序列,ggvis,lattice,和GGPLOT2可视化数据,caret 机器学习

When and how do I use Python?

Python can be used when your data analysis tasks need to consolidate Web applications, or if statistical code needs to be included in the production database. As a fully fledged programming language, it is a great tool to implement production using algorithms.

In the past, the data analysis of Python packages is an issue that has improved significantly over the years. Be sure to install numpy/scipy (scientific calculations) and pandas (data processing) so that Python can be used for data analysis. Also see matplotlib making graphics and Scikit-learn for machine learning.

Unlike R,python there is no clear "best" IDE. We recommend that you use Spyder,ipython Notebook and rodeo to see which one is best for your needs.

R and Python: the number of data sciences

If you look at recent polls and focus on the programming language for data analysis, R is often the obvious winner. If you pay particular attention to the field of Python and r data analysis, you will get a similar pattern.

Despite the above figures, more and more people are moving from R to Python. In addition, more and more people are using both. This is also in line with our recommendation to students.

If you're going to start a career in data science, it's best to be proficient in two languages. Recruitment trends show that demand for both skills is increasing and wages are far above average.

R: Pros and cons

Pro: One picture beats more than words

Visual data is often easier to understand than individual raw data. R and visualization are a perfect match. Some of the visual packages that must be seen are Ggplot2,ggvis,googlevis and rcharts.

Pro:r Eco-System

R has a rich ecosystem, which includes cutting-edge packages and active communities. Packages can be obtained on cran,bioconductor and GitHub. You can search for all packages in rdocumentation.

Pro: Using R in Data science

R is developed by statisticians for statistical experts. They can communicate ideas and ideas through R code and packages, and you don't necessarily need a computer science background. In addition, people outside academia are using r more and more.

Pro/con:r is slow.

R is developed to make life easier for statisticians, not for the life of your computer. Although R may run slowly due to poorly written code, there are multiple packages to improve R performance: Pqr,renjin and Fastr,riposte, and so on.

Con:r has a steep learning curve.

The R learning curve is complex, especially if you are a GUI from statistical analysis. If you're not familiar with it, finding a package can be time-consuming.

Python: Pros and cons

Pro:ipython Notebook

IPython notebook makes it easier to work with Python and data. You can easily share notebooks with coworkers without having to install anything. This greatly reduces the overhead of organizing code, output, and commenting files. This will allow you to spend more time doing the actual work.

Pro: Universal language

Python is a common language that is easy and intuitive. This gives it a relatively flat learning curve, which can increase the speed at which you write a program. In short, you need less time to write code!

In addition, the Python test framework is a built-in, ground-required test framework that encourages good test coverage. This will ensure that your code is reusable and reliable.

Pro: A multi-purpose language

Python brings together people from different backgrounds. As a universal, easy to understand language, statisticians can easily learn that you can build a single tool to integrate every part of your workflow.

Pro/con: Visualization

Visualization is an important criterion when choosing data analysis software. Although Python has some good visual libraries, such as Seaborn,bokeh and Pygal, there are too many options to choose from. In addition, visualization is often cumbersome compared to r, and the results are not always pleasing to the eye.

Con:python is a challenger.

Python is the challenger to R. It does not provide an alternative to the hundreds of essential R packages. Although it is catching up, it is unclear whether this will let people give up R?

Who's the winner?

It's up to you! As a data scientist, choosing the language that best meets your needs is your job. Some questions can help you:

你想解决什么问题？学习语言的净成本是什么？在你的领域中常用的工具是什么？其他可用的工具是什么和这些相关工具日常中是如何使用的？

Hope to be helpful to you!

R VS Python in Data science: The winner is ...