C++ implementation of neural networks library with Keras-like API. Contains majority of commonly used layers, losses and optimizers. Supports sequential and multi-input-output (flow) models. Supports single CPU, Multi-CPU and GPU tensor operations (using cuDNN and cuBLAS).
neural-network
gan
dcgan
autoencoder
adaptive
pix2pix
computational-graphs
cudnn
fast-neural-style
adain
neuro
neural-style-transfer
ada-in
adaptive-style-transfer
-
Updated
May 22, 2021 - C++