Computing environment · JVM
S2
A Next-Generation Data Science Toolbox
S2 is an integrated development environment of advanced analytics and data. It is orders of magnitude faster than R and Python, supports data analysis on terabytes or even petabytes of data across arrays of servers, and comes with many polished data sets and proprietary algorithms out of the box.
R alternative
Everything R does — much faster, deployable anywhere.
R code runs only inside the R environment and is very difficult to deploy elsewhere. S2 code runs in any JVM environment — and there are now 15 billion devices that run JVM, from embedded devices to space rockets.
Python replacement
Scripting and tensor programming, without the deployment nightmare.
S2 scripting compiles to Java bytecode and is orders of magnitude faster than Python. It runs on any JVM device by copy-and-pasting jars — no versioning of dependencies, no assorted FORTRAN/C/C++ libraries.
Numeric computing environment
01S2 is an IDE for coding numerical algorithms. Start with 1+1 — and with 1+1 working, you can do pretty much anything in numerical programming, which is just a more complicated series of many 1+1. S2 has the fastest linear algebra package in the Java world, probably.
Linear algebra
02 · matrix operations and beyondSee the full collection of linear algebra algorithms in the NM Dev packages or try them out yourself in the S2 IDE. For example, to do matrix operations, one can simply write
Plotting
03 · dozens of chart types
Statistics
04A very comprehensive statistics package: almost all types of linear regressions and their statistics — OLS, GLM and logistics — plus time series analysis, random number generation, distributions, and many more.
Optimization
05Solvers are the foundation of future mathematics — the core of AI and Big Data analytics. S2 supports a full suite of all known standard optimization algorithms.
06 · Machine learning
Train a neural network in a few lines.
A simple script trains a neural network to learn the Black-Scholes formula from a data set of stock prices and option prices. It converges in a few hundred epochs.
07 · Scripting
A steel plant's scheduling system in 12 lines.
Read the job data, read the machine data, schedule the jobs to the machines, plot the job-shop schedules to maximize utilization — all in one S2 script. The same code deploys on S2, in a stand-alone application, or on a cloud using REST.
08 · Array / tensor programming
Parallel execution, the way numpy and pandas do it.
S2 supports array/tensor programming for parallelization — apply a formula concurrently to all rows of a table, and dissect, slice, cut, dice, sample and massage data frames using ND4J. S2 also handles terabytes or petabytes of data across arrays of machines with map-reduce in a simple script.
Industrial partners
A one-stop shop of algorithms and dataAlgoQuant
A large library of financial analytics with hundreds of functions, plus well-cleaned, professionally maintained data for US and China equities. Write a moving-average crossover strategy in a few lines, plug it into the AlgoQuant framework for backtesting, and use the suite of analysis and reporting tools.
SuperCurve
A fixed income data firm in China selling high-quality bond data and analytics. Retrieve China bond data in S2 using the SuperCurve API and fit a zero-coupon yield curve in two lines of code — then price any fixed income instrument on that date.
Licensing
Community Edition
Free to use — try it without registering an account (your work won't be saved). Registration is free. Tell us what you think, bugs and feature requests, in our forum.
Launch S2 →Enterprise Edition
Increase S2's computational power, run a private and secured server (cluster), or co-develop and customize S2 with us.
Contact sales →Third-party vendors
Host your data and/or algorithms and analytics on S2 and reach every S2 user.
Contact us →