About: A Python library that allows you to define, optimize, and evaluate mathematical expressions involving multi-dimensional arrays efficiently. Dynamically generates CPU and GPU modules for good performance. Deep Learning Tutorials illustrate deep learning with Theano.

Theano 1.0.2 (23rd of May, 2018)

This is a maintenance release of Theano, version 1.0.2, with no new features, but some important bug fixes.

We recommend that everybody update to this version.

Highlights (since 1.0.1):

• Theano should work under PyPy now (this is experimental).
• Update for cuDNN 7.1 RNN API changes.
• Fix for a crash related to mixed dtypes with cuDNN convolutions.
• MAGMA should work in more cases without manual config.
• Handle reductions with non-default accumulator dtype better on the GPU.
• Improvements to the test suite so that it fails less often due to random chance.

A total of 6 people contributed to this release since 1.0.1:

• Frederic Bastien
• Steven Bocco
• Jon Haygood
• Arnaud Bergeron
• Jordan Melendez
• Desiree Vogt-Lee
• Garming Sam
• Pascal Lamblin
• Vincent Dumoulin
• Glexin
• Simon Lefrancois

About: Somoclu is a massively parallel implementation of self-organizing 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.

• New: A Makefile for mingw to build on Windows.
• Changed: PR #94 added a much more efficient sparse kernel.
• Changed: boilerplate code for Julia greatly improved.
• Changed: Code cleanup, pre-processor macros simplified.
• Changed: Adapted to Seaborn API changes in plotting heatmaps.

About: CURFIL uses NVIDIA CUDA to accelerate random forest training and prediction for RGB and RGB-D images. It focuses on image labelling tasks, such as image segmentation or classification applications. CURFIL allows to search for optimal hyper-parameter configurations (e.g. using the hyperopt) package) by massively decreasing training time.

Initial Announcement on mloss.org.

About: Caffe aims to provide computer vision scientists with a clean, modifiable implementation of state-of-the-art deep learning algorithms. We believe that Caffe is the fastest available GPU CNN implementation. Caffe also provides seamless switching between CPU and GPU, which allows one to train models with fast GPUs and then deploy them on non-GPU clusters. Even in CPU mode, computing predictions on an image takes only 20 ms (in batch mode).

LOTS of stuff: https://github.com/BVLC/caffe/releases/tag/v0.9999

About: GPUML is a library that provides a C/C++ and MATLAB interface for speeding up the computation of the weighted kernel summation and kernel matrix construction on GPU. These computations occur commonly in several machine learning algorithms like kernel density estimation, kernel regression, kernel PCA, etc.

Initial Announcement on mloss.org.