About: A Graphical Tool for Designing and Training Deep Neural Networks Changes:

About: Somoclu is a massively parallel implementation of selforganizing maps. It relies on OpenMP for multicore execution, MPI for distributing the workload, and it can be accelerated by CUDA on a GPU cluster. A sparse kernel is also included, which is useful for training maps on vector spaces generated in text mining processes. Apart from a command line interface, Python, Julia, R, and MATLAB are supported. Changes:

About: MLweb is an open source project that aims at bringing machine learning capabilities into web pages and web applications, while maintaining all computations on the client side. It includes (i) a javascript library to enable scientific computing within web pages, (ii) a javascript library implementing machine learning algorithms for classification, regression, clustering and dimensionality reduction, (iii) a web application providing a matlablike development environment. Changes:

About: Armadillo is a high quality C++ linear algebra library, aiming towards a good balance between speed and ease of use. The function syntax is deliberately similar to MATLAB. Useful for algorithm development directly in C++, or quick conversion of research code into production environments (eg. software & hardware products). Changes:

About: DataDeps is a package for simplifying the management of data in your julia application. In particular it is designed to make getting static data from some server into the local machine, and making programs know where that data is trivial. Changes:Initial Announcement on mloss.org.

About: This project is a C++ toolkit containing machine learning algorithms and tools for creating complex software in C++ to solve real world problems. Changes:This release removes the need for Boost.Python when using dlib via Python. This makes compiling the Python interface to dlib much easier as there are now no external dependencies.

About: The Weka workbench contains a collection of visualization tools and algorithms for data analysis and predictive modelling, together with graphical user interfaces for easy access to this [...] Changes:This release include a lot of bug fixes and improvements. Some of these are detailed at http://jira.pentaho.com/projects/DATAMINING/issues/DATAMINING771 As usual, for a complete list of changes refer to the changelogs.

About: Operator Discretization Library (ODL) is a Python library that enables research in inverse problems on realistic or real data. Changes:Initial Announcement on mloss.org.

About: A Spark package implementing algorithms for learning from crowdsourced big data. Changes:Changes:  Minor improvements in code and documentation

About: A Python library that allows you to define, optimize, and evaluate mathematical expressions involving multidimensional arrays efficiently. Dynamically generates CPU and GPU modules for good performance. Deep Learning Tutorials illustrate deep learning with Theano. Changes:Theano 1.0.1 (6th of December, 2017)This is a maintenance release of Theano, version Highlights (since 1.0.0):

About: A WEKA package for analyzing emotion and sentiment of tweets. Changes:Initial Announcement on mloss.org.

About: The Accord.NET Framework is a .NET machine learning framework combined with audio and image processing libraries completely written in C#. It is a complete framework for building productiongrade computer vision, computer audition, signal processing and statistics applications even for commercial use. A comprehensive set of sample applications provide a fast start to get up and running quickly, and an extensive online documentation helps fill in the details. Changes:For a complete list of changes, please see the full release notes at the release details page at: https://github.com/accordnet/framework/releases/tag/v3.8.0

About: A Java framework for statistical analysis and classification of biological sequences Changes:New classes and packages:
New features and improvements:

About: A noniterative, incremental and hyperparameterfree learning method for onelayer feedforward neural networks without hidden layers. This method efficiently obtains the optimal parameters of the network, regardless of whether the data contains a greater number of samples than variables or vice versa. It does this by using a square loss function that measures errors before the output activation functions and scales them by the slope of these functions at each data point. The outcome is a system of linear equations that obtain the network's weights and that is further transformed using Singular Value Decomposition. Changes:Initial Announcement on mloss.org.

About: A noniterative learning method for onelayer (no hidden layer) neural networks, where the weights can be calculated in a closedform manner, thereby avoiding low convergence rate and also hyperparameter tuning. The proposed learning method, LANNSVD in short, presents a good computational efficiency for largescale data analytic. Changes:Initial Announcement on mloss.org.

About: MSVMpack is a Multiclass Support Vector Machine (MSVM) package. It is dedicated to SVMs which can handle more than two classes without relying on decomposition methods and implements the four MSVM models from the literature: Weston and Watkins MSVM, Crammer and Singer MSVM, Lee, Lin and Wahba MSVM, and the MSVM2 of Guermeur and Monfrini. Changes:

About: OpenNN is a software library written in C++ for advanced analytics. It implements neural networks, the most successful machine learning method. The library has been designed to learn from both data sets and mathematical models. Changes:New algorithms, correction of bugs.

About: A mutual information library for C and Mex bindings for MATLAB. Aimed at feature selection, and provides simple methods to calculate mutual information, conditional mutual information, entropy, conditional entropy, Renyi entropy/mutual information, and weighted variants of Shannon entropies/mutual informations. Works with discrete distributions, and expects column vectors of features. Changes:Fixed a Windows compilation bug. MIToolbox v3 should now compile using Visual Studio.

About: The new R package opusminer provides an R interface to the OPUS Miner algorithm (implemented in C++) for finding the key associations in transaction data efficiently, in the form of selfsufficient itemsets, using either leverage or lift. Changes:Initial Announcement on mloss.org.

About: The proposed hashing algorithm leverages the bootstrap sampling idea and integrates it with PCA, resulting in a new projection method called Bagging PCA Hashing. Changes:Initial Announcement on mloss.org.
