DaNNet is a C++ deep neural network library using the Armadillo library as a base. It is intended to be a small and easy to use framework with no other dependencies than Armadillo. It uses independent layer-wise optimization giving you full flexibility to train your network.
Features
- Layers: Input, Dense, Convolution, Pool, Batch norm, Dropout, Cost
- Activations: Sigmoid, tanh, ReLU, LReLU, softplus, softmax
- Independent layer-wise optimizations
- Optimizers; SGD, Nesterov, ADAM, RMSprop, ADAdelta, ADAgrad, ADAmax
- Regularizations: LASSO, ridge, elastic net
- MNIST import support
License
Apache License V2.0Follow DaNNet
You Might Also Like
Rate This Project
Login To Rate This Project
User Reviews
Be the first to post a review of DaNNet!