An implementation of "MixHop: Higher-Order Graph Convolutional Architectures via Sparsified Neighborhood Mixing" (ICML 2019).

Real time monaural source separation base on fully convolutional neural network operates on Time-frequency domain.

Time Series Feature Extraction using Deep Learning

Pytorch implementation of Hebbian learning algorithms to train deep convolutional neural networks.

Lego parts classifier using deep neural networks

A naive (very simple!) implementation of a convolutional neural network

Deep Learning (Keras) Models Deployment using SQL databases

Pytorch implementation of Hebbian learning algorithms to train deep convolutional neural networks.

Hands-on in-person workshop for Deep Learning with TensorFlow

Convolutional encoder and a Viterbi decoder written in Golang

Visualize feature maps in convolutional neural networks.

Convolutional autoencoder for encoding/decoding RGB images in TensorFlow with high compression ratio

Deep Learning (PyTorch) Models Deployment using SQL databases

AI High-Performance Solution on FPGA

Some neural networks code and notes I've been doing for Andrew Ng's Deep Learning course.

Japanese Handwritten Character Recognition using DropBlock Regulzarization

a light-weight convolutional neural network based on c++ 11

Mokka is a minimal Inference Engine for Dense and Convolutional 2D Layer Neural Networks. Written on a single C++ header, it uses AVX2

Neural network for words recognition

Lightweight, zero dependency, machine learning library.