zshwuhan Goto Github PK

neural_stpp

Deep generative modeling for time-stamped heterogeneous data, enabling high-fidelity models for a large variety of spatio-temporal domains.

neurec

Next RecSys Library

news-classifier

给定训练新闻数据集，可以对输入的测试新闻进行自动分类识别

news-popularity-prediction

A set of methods that predict the future values of popularity indices for news posts using a variety of features.

news_event_evolution

Event evolution detection based on Graph network model

nimfa

Nimfa - A Python module for nonnegative matrix factorization

nips14-ssl

Code for reproducing results of NIPS 2014 paper "Semi-Supervised Learning with Deep Generative Models"

nips2015_vrnn

nips2016

Code used for results provided at NIPS 2016

nlp_dl_lecture_note

nlpir

nmf-toolbox

Toolbox for performing Non-negative Matrix Factorization (NMF) and several variants

nmflib

nonnegative matrix factorization and some variants in numpy

nmflibrary

MATLAB library for non-negative matrix factorization (NMF): Version 1.8.1

nmflogreg

Coupled Nonnegative Matrix Factorization and Logistic Regression

nmt.hybrid

State-of-the-art Neural Machine Translation Codebase including Hybrid Word-character Models

nmtf

An implementation of the paper "Nonnegative Matrix Tri-Factorization with Graph Regularization for Community Detection in Social Networks"

nmtf-1

Non-Negative Matrix Tri-Factorization for Co-clustering

nn_pred

Deep learning for time-series prediction.

nncf

Code for paper "On Sampling Strategies for Neural Network-based Collaborative Filtering"

nnformll

Neural Network Models for Multi-label learning

node2vec

nolineartimeseriesanalysis

The codes in the toolbox can be used to perform nonlinear time series analysis on single(or multi) channel data. This is done by mapping the single channel data to phase space representation using Taken's embedding theorem (compute_psv.m). The parameters - optimal delay and dimension are estimated using first minimum of MI (compute_tau.m) and FNN method (compute_dim) respectively. The recurrence network can be constructed from the phase space vector using ComputeRecurrenceNetwork_ANN.m or ComputeRecurrenceNetwork_fixedRR.m. The topology of the RN can be further analysed using graph theoreticl quantifiers (you need BCT toolbox for this). One can also compute the complexity-entrropy plane using get_mpr_complexity.m for which the ordinal patterns are computed using get_ordinal_pattern_dist.m (see the function descp for more details). Also, the tool box contains python codes to generate variety of uni(or multi) variate surrogate data.

nonlinearorthogonalnmf

A Nonlinear Orthogonal Non-Negative Matrix Factorization Approach to Subspace Clustering

notes-on-dirichlet-processes

:game_die: IPython notebooks explaining Dirichlet Processes, HDPs, and Latent Dirichlet Allocation

npb_phenotyping

Bayesian Nonparametric Tensor Factorization for Phenotyping

npmm

A nonparametric model for online topic discovery with word embeddings

ntu-ml-2014fall

opinion-mining-system

新闻评论观点挖掘系统，粗粒度的分析出新闻网评观点的倾向和走势

optics-automatic-clustering

automatic hierarchical clustering of OPTICS reachability plots implemented in Python