dongrongliang Goto Github PK

aistock

autokeras

accessible AutoML for deep learning.

bilateral

Bilateral Filter

chinese-word-vectors

100+ Chinese Word Vectors 上百种预训练中文词向量

deepmind-research

This repository contains implementations and illustrative code to accompany DeepMind publications

diffusion

Denoising Diffusion Probabilistic Models

easoncodeshare

just for share my code

elasticsearch-feature-vector-scoring

A simple scoring plugin for vector in Elasticsearch.

fast-adaptive-bilateral-filtering

Matlab code for "Fast Adaptive Bilateral Filtering", IEEE TIP

fast-neural-style

Feedforward style transfer

fast-neural-style-1

Implementation of "Perceptual Losses for Real-Time Style Transfer and Super-Resolution" in Keras

fastphotostyle

Style transfer, deep learning, feature transform

foolnltk

A Chinese Nature Language Toolkit

ganotebooks

wgan, wgan2(improved, gp), infogan, and dcgan implementation in lasagne, keras, pytorch

generative-latent-optimization-tensorflow

Tensorflow implementation of Generative Latent Optimization (GLO) proposed by Facebook AI Research

gitskills

hdrnet

An implementation of 'Deep Bilateral Learning for Real-Time Image Enhancement', SIGGRAPH 2017

httpproxymiddleware

A middleware for scrapy. Used to change HTTP proxy from time to time.

image-style-transform

In our project, we use the perceptual loss functions for training feed-forward networks instead of the traditional per-pixel loss between the output and ground-truth images (style image and content image) for image style transformation. The method we use can do the transformation almost in real- time via GPU. We divide our neural network into two parts, the image transform net and the loss net. For the image transform net we compare the result of two image transform net, with or without residual blocks. By the method of extracting high-level features from pre-trained loss net, we can optimize the quality of images generated. Our loss function contains three different parts, that is, style reconstruction loss, feature reconstruction loss and total variation loss for style penalty, content penalty and spatial smoothness in the output image separately. Also, we do a multiple style transfer and make some improvements for the quality of the transformed image.

image-super-resolution

Implementation of Super Resolution CNN in Keras.

improved_wgan_training

Code for reproducing experiments in "Improved Training of Wasserstein GANs"

imutils

A series of convenience functions to make basic image processing operations such as translation, rotation, resizing, skeletonization, and displaying Matplotlib images easier with OpenCV and Python.

instcolorization

kafkaconsumerlag

keras

Deep Learning library for Python. Runs on TensorFlow, Theano, or CNTK.

keras-coordconv

Keras implementation of CoordConv for all Convolution layers

keras-gan

Keras implementations of Generative Adversarial Networks.

keyword-extraction-datasets

Different datasets for developing and testing keyword extraction algorithms

latexify_py

Generates LaTeX math description from Python functions.

learngit