Scientific Software

Coq is a formal proof management system. It provides a formal language to write mathematical definitions, executable algorithms and theorems together with an environment for semi-interactive development of machine-checked proofs. Typical applications include the certification of properties of programming languages (e.g. the CompCert compiler certification project, or the Bedrock verified low-level programming library), the formalization of mathematics (e.g. the full formalization of the Feit-Thompson theorem or homotopy type theory) and teaching.

A number of libraries and packages in Python, C/C++, Lua for machine learning and artificial intelligence by the Facebook Research group.

fastText is a library for efficient learning of word representations and sentence classification. fastText builds on modern Mac OS and Linux distributions. Since it uses C++11 features, it requires a compiler with good C++11 support. It is developed by Facebook Research.

Descartes Labs is excited to release GeoVisual Search. We’ve used the power of our geospatial platform to process public and commercial satellite imagery, detect visual similarities between scenes, and apply machine learning to recognize different types of objects across the globe.

We are launching GeoVisual Search on three different imagery catalogs, showcasing public and commercial imagery, multiple resolutions, and both national and global scale search. You can explore each of these, today.

The Committee on Earth Observation Satellites (CEOS) has long recognized a need for data processing infrastructure to support Earth science objectives in developing countries. Forest preservation initiatives, carbon measurement initiatives, water management and agricultural monitoring are just few examples of causes that can benefit greatly from remote sensing data. Currently, however, many developing nations lack the in-country expertise and computational infrastructure to utilize remote sensing data. The CEOS Data Cube Platform provides a flexible model to address these needs. The CEOS Data Cube Platform is a data processing platform for Earth science data, with a focus on remote-sensing data. The platform provides a data ingestion framework that includes support for automated ingestion of a wide variety of remote sensing data products. The data products are ingested into an N-dimensional data array that abstracts away management of distinct acquisitions. The platform has a tiered API for data processing and a data/application platform layer for higher-level access.

Thousands of users rely on Stan for statistical modeling, data analysis, and prediction in the social, biological, and physical sciences, engineering, and business.

Users specify log density functions in Stan’s probabilistic programming language and get:
- full Bayesian statistical inference with MCMC sampling (NUTS, HMC)
- approximate Bayesian inference with variational inference (ADVI)
- penalized maximum likelihood estimation with optimization (L-BFGS)

Stan’s math library provides differentiable probability functions & linear algebra (C++ autodiff). Additional R packages provide expression-based linear modeling, posterior visualization, and leave-one-out cross-validation.

Stan interfaces with the most popular data analysis languages (R, Python, shell, MATLAB, Julia, Stata) and runs on all major platforms (Linux, Mac, Windows).

Edward is a Python library for probabilistic modeling, inference, and criticism. It is a testbed for fast experimentation and research with probabilistic models, ranging from classical hierarchical models on small data sets to complex deep probabilistic models on large data sets. Edward fuses three fields: Bayesian statistics and machine learning, deep learning, and probabilistic programming.

Edward is built on top of TensorFlow. It enables features such as computational graphs, distributed training, CPU/GPU integration, automatic differentiation, and visualization with TensorBoard.

Mathematical optimization is a well-studied language of expressing solutions to many real-life problems that come up in machine learning and many other fields such as mechanics, economics, EE, operations research, control engineering, geophysics, and molecular modeling. As we build our machine learning systems to interact with real data from these fields, we often cannot (but sometimes can) simply ``learn away'' the optimization sub-problems by adding more layers in our network. Well-defined optimization problems may be added if you have a thorough understanding of your feature space, but oftentimes we don't have this understanding and resort to automatic feature learning for our tasks.

Until this repository, no modern deep learning library has provided a way of adding a learnable optimization layer (other than simply unrolling an optimization procedure, which is inefficient and inexact) into our model formulation that we can quickly try to see if it's a nice way of expressing our data.

See our paper OptNet: Differentiable Optimization as a Layer in Neural Networks and code at locuslab/optnet if you are interested in learning more about our initial exploration in this space of automatically learning quadratic program layers for signal denoising and sudoku.

Scikit-plot is an intuitive library to add plotting functionality to scikit-learn objects.

Scikit-plot is the result of an unartistic data scientist's dreadful realization that visualization is one of the most crucial components in the data science process, not just a mere afterthought.

Gaining insights is simply a lot easier when you're looking at a colored heatmap of a confusion matrix complete with class labels rather than a single-line dump of numbers enclosed in brackets. Besides, if you ever need to present your results to someone (virtually any time anybody hires you to do data science), you show them visualizations, not a bunch of numbers in Excel.

That said, there are a number of visualizations that frequently pop up in machine learning. Scikit-plot is a humble attempt to provide aesthetically-challenged programmers (such as myself) the opportunity to generate quick and beautiful graphs and plots with as little boilerplate as possible.

We speed up the image generation algorithm of PixelCNN++ by avoiding redundant computation through caching. Naive generation discards computation that can be re-used and performs additional computation that will not be used to generate a particular pixel. Naive generation can take up to 11 minutes to generate 16 32-by-32 images on a Tesla K40 GPU. By re-using previous computation and only performing the minimum amount of computation required, we achieve up to a 183 times speedup over the naive generation algorithm. We have tested our code with Python 3 and TensorFlow 1.0. You may need to make small changes for other versions of Python or TensorFlow.

Prophet is a procedure for forecasting time series data developed by Facebook’s Core Data Science team. It is based on an additive model where non-linear trends are fit with yearly and weekly seasonality, plus holidays. It works best with daily periodicity data with at least one year of historical data. Prophet is robust to missing data, shifts in the trend, and large outliers.

Prophet is open source software and available for download on CRAN and PyPI.

Python implementations of various Machine Learning models and algorithms from scratch.

Dat is a grant-funded, open-source, decentralized data sharing tool for efficiently versioning and syncing changes to data.

Dat can be used to version data locally, or to share and sync data over the internet. Dat includes an optional peer-to-peer distribution system, meaning that the more widely that a dataset is shared, the faster it is for users to retrieve or sync a copy, and the more redundant that the dataset’s availability becomes.

By building tools to build and share data pipelines, we aim to bring to data a style of collaboration similar to what Git brings to source code. Dat is designed as a general-purpose tool for any data on the Web, with our main priority being to ensure scientific data can be more easily published and archived. Dat is fully open source and is built using JavaScript, Node.js and Electron.

Web2py is a free open source full-stack framework for rapid development of fast, scalable, secure and portable database-driven web-based applications. Written and programmable in Python. Web2py is created by a community of professionals and University professors in Computer Science and Software Engineering.

Features incude:
- Always backward compatible. We have not broken backward compatibility since version 1.0 in 2007, and we pledge not to break it in the future.
- Easy to run. It requires no installation and no configuration.
- Runs on Windows, Mac, Unix/Linux, Google App Engine, Amazon EC2, and almost any web hosting via Python 2.5/2.6/2.7/pypy, or Java with Jython.
- Runs with Apache, Lighttpd, Cherokee and almost any other web server via CGI, FastCGI, WSGI, mod_proxy, and/or mod_python. It can embed third party WSGI apps and middleware.

MALLET is a Java-based package for statistical natural language processing, document classification, clustering, topic modeling, information extraction, and other machine learning applications to text.

MALLET includes sophisticated tools for document classification: efficient routines for converting text to "features", a wide variety of algorithms (including Naïve Bayes, Maximum Entropy, and Decision Trees), and code for evaluating classifier performance using several commonly used metrics.

The lda2vec model tries to mix the best parts of word2vec and LDA into a single framework. word2vec captures powerful relationships between words, but the resulting vectors are largely uninterpretable and don't represent documents. LDA on the other hand is quite interpretable by humans, but doesn't model local word relationships like word2vec. We build a model that builds both word and document topics, makes them interpreable, makes topics over clients, times, and documents, and makes them supervised topics.

The Palm Generator is a Three.js module to generate palm trees. It follows a model that describes the arrangement of leaves on a plant stem, called phyllotaxis. You can find information about the usage and the license on GitHub. You can create your palm using the PalmGenerator Online Editor.

Basin and Landscape Dynamics (Badlands) is a parallel TIN-based landscape evolution model, built to simulate topography development at various space and time scales. The model is capable of simulating hillslope processes, fluvial incision (erosion/transport/deposition), spatially and temporally varying geodynamic (3D displacements) and climatic forces which can be used to simulate changes in base level, as well as effects of climate changes or sea-level fluctuations.

Badlands is under active development by Dr Tristan Salles within the EarthByte group.

GPlates is desktop software for the interactive visualisation of plate-tectonics.

GPlates offers a novel combination of interactive plate-tectonic reconstructions, geographic information system (GIS) functionality and raster data visualisation. GPlates enables both the visualisation and the manipulation of plate-tectonic reconstructions and associated data through geological time. GPlates runs on Windows, Linux and MacOS X. GPlates has an online user manual.

A PHP wrapper for the Stanford Natural Language Processing library. Supports POSTagger and CRFClassifier. Loads automatically the right packages and detects the language of the given text.