PyDTNN: Python Distributed Training of Neural Networks
PyDTNN is a light-weight framework developed at Universitat Jaume I for distributed Deep Learning training and inference that offers an initial starting point for interaction with deep neural networks.
The convWinograd library is an implementation of the Winograd-based convolution transform, which can be used within PyDTNN.
The convDirect library provides different implementations of the direct convolution operation. This operation can work within PyDTNN.
The convGemm library performs the convolution operation using an implicit im2row or im2col over a GEMM operation with matrices in either the NHWC or NCHW format, respectively. This operation can work within PyDTNN.
PMLib: Power Measurement Library
PMLib is a modular open-source power measurement library to investigate power and energy usage of high-performance computing applications.
GrPPI: Generic Reusable Parallel Pattern Interface
GrPPI is an open-source generic and reusable parallel pattern programming interface developed at University Carlos III of Madrid. GrPPI accommodates a layer between developers and existing parallel programming frameworks for multi-core processors, such as ISO C++ Threads, OpenMP, Intel TBB, and FastFlow.