English»Software»Machine Learning»Deep learning»Frameworks»Efficiency | searchivarius.org
log in | about 

Block-sparse GPU kernels   - highly-optimized GPU kernels for an underexplored class of neural network architectures: networks with block-sparse weights.
DarkNet   - A neural network framework written in C and CUDA.
FeatherCNN   - a high performance inference engine for convolutional neural networks.
SOD   - An Embedded Computer Vision & Machine Learning Library SOD.
Tensor comprehensions   - efficient deep learning seamlessly integrated into Pytorch.
TensorNetwork   - A library for easy and efficient manipulation of tensor networks.
Thinc   - Practical Machine Learning for NLP in Python. t features a battle-tested linear model designed for large sparse learning problems, and a flexible neural network model with efficient CPU usage.
Tiny-DNN   - a C++11 implementation of deep learning. It is suitable for deep learning on limited computational resource, embedded systems and IoT devices.
TorchGPipe   - A GPipe implementation in PyTorch
tvm   - Open deep learning compiler stack for cpu, gpu and specialized accelerators.
VectorFlow   - a minimalist neural network library optimized for sparse data and single machine environments. It is described in this blog post.