Slant community chat on Discourse
Pro Rich set of libraries and extensions
Reduces the lines of code required to get stuff done.
Pro Full programming language, unlike R
Allows for more rapid deployment of solutions.
Con Lack of statistical and machine learning algorithms
Compared to programming languages like R, which were created for Data Science, Python is missing a lot of old and new statistical approaches.
Pro Nice regular syntax
Julia code is easy to read and avoid a lot of unnecessary special symbols and fluff. It uses newline to end statements and "end" to end blocks so there is no need for lots of semicolons and curly braces. It is regular in that unless it is a variable assignment, function name always comes...
Con Young language with limited support
Julia was released in 2012. Due to its short existence, there is a limited amount of support for the language. Very few libraries exist as of yet, and the community is still quite small (though growing quickly).
Pro Rich choice of high quality libraries
Be it data manipulation or plotting there is always a wide variety at CRAN available.
Pro Great RStudio IDE
It's free, lightweight, and easy to use.
Con The library ecosystem (mostly CRAN but also github, BioConductor & others) is rather fragmented
There are over 10000 R packages and rising and there are many packages with overlapping functionality. It's not always clear which package is "best" for your needs. The situation is getting better thanks to Hadley Wickham who has rewritten a lot of core functionality from R base &
Pro Matlab compatibility
The languages are so similar that it's easy to port Matlab libraries.
Pro Excellent for symbolic computing
e.g. solving differential equations analytically.
Con Not free or open source, expensive for commercial non-academic use
Nuance: you can experiment with it for free now in the Wolfram cloud.
Pro Maximum performance
Fortran compilers are very mature--not even C is faster.
Con No library ecosystem for modern data science: exploratory data analysis, statistics, machine learning
Fortran is good for pure number crunching on supercomputers. The available open source libraries are mostly in domains like physics or climate science and they are mostly about things like simulations or partial differential equations solving. So it's more suitable for going from models to dat...
Comments
Welcome to the Slant Community
Stop spending hours researching
Slant is written by a community helping you be informed. Let us know what you’re passionate about to get an awesome personalized feed.