ai42 Goto Github PK

fork-om3d

3D Object Manipulation in a Single Photograph

gan_deeplearning4j

Automatic feature engineering using Generative Adversarial Networks using Deeplearning4j and Apache Spark.

general_deep_q_rl

General deep reinforcement learning framework

gocl

Go OpenCL (GOCL) Binding

gopencl

Go to OpenCL compiler

gorgonia

Gorgonia is a library that helps facilitate machine learning in Go.

hanlp

自然语言处理 中文分词 词性标注 命名实体识别 依存句法分析 关键词提取 自动摘要 短语提取 拼音 简繁转换

have-fun-with-machine-learning

An absolute beginner's guide to Machine Learning and Image Classification with Neural Networks

human_level_control_through_deep_reinforcement_learning

Deepmind's Deep Q Learning on Atari

imagenet-api

基于Keras+Tensorflow搭建，提供ResNet50神经网络的图片分类平台。

jsimgproc

Jethro's Image Processing Software.Super-Resolution Reconstruction is the most import feature in this software. It’s achieved by the algorithm Single-Image Super Resolution, based on theory of Compressed Sensing. This algorithm uses OMP (Orthogonal Matching Pursuit) to compute the sparse coding of images, K-SVD (K-means Singular Value Decomposition) to learn over-complete dictionary. With the collaboration between OMP and over-complete dictionary, low resolution image can be scaled-up and its high frequency information (texture, edge) can be recovered at some level. The result is better than simple interpolation, eg, linear interpolation, bi-cubic interpolation, etc. This algorithm is implemented using C++&OpenCV. The other features of this software are also implemented using C++&OpenCV. The GUI is designed using QT.

keras

Deep Learning library for Python. Convnets, recurrent neural networks, and more. Runs on Theano and TensorFlow.

keras-frcnn-web

使用keras建立faster rcnn网络，训练完成后，将识别接口释放为rest，方便后端调用

lasagne

Lightweight library to build and train neural networks in Theano

learningsimplealgorithms

We present an approach for learning simple algorithms such as copying, multi-digit addition and single digit multiplication directly from examples. Our framework consists of a set of interfaces, accessed by a controller. Typical interfaces are 1-D tapes or 2-D grids that hold the input and output data. For the controller, we explore a range of neural network-based models which vary in their ability to abstract the underlying algorithm from training instances and generalize to test examples with many thousands of digits. The controller is trained using Q-learning with several enhancements and we show that the bottleneck is in the capabilities of the controller rather than in the search incurred by Q-learning.

libnd4j

librec

LibRec: A Leading Java Library for Recommender Systems, see

memn2n

End-To-End Memory Networks in Theano

memnn

Theano implementation of Memory Networks

minerva

Minerva: a fast and flexible tool for deep learning on multi-GPU. It provides ndarray programming interface, just like Numpy. Python bindings and C++ bindings are both available. The resulting code can be run on CPU or GPU. Multi-GPU support is very easy.

model-sanctuary

Safe space to observe and interact with existing neural net models.

mshadow

Matrix Shadow:Lightweight CPU/GPU Matrix and Tensor Template Library in C++/CUDA for (Deep) Machine Learning

mxnet

Lightweight, Portable, Flexible Distributed/Mobile Deep Learning with Dynamic, Mutation-aware Dataflow Dep Scheduler; for Python, R, Julia, Go, Javascript and more

mxnet-gtc-tutorial

MXNet Tutorial for NVidia GTC 2016.

mxnet.cpp

CPP wrapper for MXNet interface

nd4j

nd4j site

nd4j-kernels

nematus

neural-style

Torch implementation of neural style algorithm

neuralart

An implementation of the paper 'A Neural Algorithm of Artistic Style'.