sunyong2016

academicpages.github.io

Github Pages template for academic personal websites, forked from mmistakes/minimal-mistakes

ANFIS

Fuzzy Q learning in Tensorflow , Training an ANFIS.

DeepRL-Chinese

PostGIS-Examples

常用的PostGIS的一些函数使用方法以及自定义函数，比如：pg连接oracle、pg连接sqlserver、导出csv、导入csv、查询XX米范围内数据、模糊查询、热力图聚合、生成扇形、生成栅格、生成泰森多边形、生成蜂巢图、裁剪多边形、计算点所在区域、路径分析、近邻计算等

pytorch_cluster

PyTorch Extension Library of Optimized Graph Cluster Algorithms

Semantic-Parsing

shp2geosparql

Java console tool to convert geometries from shape files to RDF using GeoSPARQL OGC standard . You can extract geometries and spatial relations of a SHP and convert them to RDF in WGS84 Reference System. Geometry2RDF: https://github.com/boricles/geometry2rdf was taken as the base of this develpment.

SKS

A high performance geospatial database that supports spatial keyword search.

TTMF

TTMF: A Triple Trustworthiness Measurement Frame for Knowledge Graphs

sunyong2016/Averaged-DQN

Averaged-DQN implemented by Chainer

sunyong2016/BITS-F312

Neural Networks and Fuzzy Logic - BITS Goa

sunyong2016/Building-Web-and-Mobile-ArcGIS-Server-Applications-with-JavaScript-Second-Edition

Building Web and Mobile ArcGIS Server Applications with JavaScript – Second Edition, published by Packt

sunyong2016/calcite-maps

A Bootstrap theme for designing, styling and creating modern map apps.

sunyong2016/contextual-bandits-recommender

Implementing LinUCB and HybridLinUCB in Python.

sunyong2016/Coursera-ML-AndrewNg-Notes

吴恩达老师的机器学习课程个人笔记

sunyong2016/data

sunyong2016/David-Silver-Reinforcement-learning

Notes for the Reinforcement Learning course by David Silver along with implementation of various algorithms.

sunyong2016/Deep-Reinforcement-Learning-in-Large-Discrete-Action-Spaces

Implementation of the algorithm in Python 3, TensorFlow and OpenAI Gym

sunyong2016/Deep_Multimodal_Embedding

Reproduction of 'Deep Multimodal Embedding: Manipulating Novel Objects with Point-clouds, Language and Trajectories'

sunyong2016/deeplearning_ai_books

deeplearning.ai（吴恩达老师的深度学习课程笔记及资源）

sunyong2016/DeepST

Deep Learning for Spatio-Temporal Data

sunyong2016/DensityPeakCluster

A cluster framework for 'Clustering by fast search and find of density peaks' in science 2014.

sunyong2016/geoQAS

A geo question answering system using strabon, word2vec, wikipedia, OpenStreetMap, gensim, nltk

sunyong2016/interpolation

Spatial interpolation methods.

sunyong2016/Machine-Learning-for-Cyber-Security

Curated list of tools and resources related to the use of machine learning for cyber security

sunyong2016/mapping

Some tools for mapping are stored here.

sunyong2016/Microsoft_taxi_data_visualization

🚕 https://www.microsoft.com/en-us/research/publication/t-drive-trajectory-data-sample/

sunyong2016/Movie_recommendation

Collaborative filtering (CF) is a technique used by recommender systems. Here a c# vs project to suggest Movie to user based on Collaborative filtering (CF)

sunyong2016/polarhub

The Polar Hub on Jekyll

sunyong2016/pro-deep-learning-w-tensorflow

Source code for 'Pro Deep Learning with TensorFlow' by Santanu Pattanayak

sunyong2016/Programming-Collective-Intelligence-Source-Code

集体智慧编程源代码

sunyong2016/py-gdalogr-cookbook

A cookbook full of recipes for using the Python GDAL/OGR bindings.

sunyong2016/python-scraping

Code samples from the book Web Scraping with Python http://shop.oreilly.com/product/0636920034391.do

sunyong2016/reinforcement_learning

Implementation of selected reinforcement learning algorithms in Tensorflow. A3C, DDPG, REINFORCE, DQN, etc.

sunyong2016/spatialanalysis

sunyong2016/TensorFlow_learning_notes

tensorflow学习笔记，来源于电子书：《Tensorflow实战Google深度学习框架》

sunyong2016/Text-Mining-with-Machine-Learning-and-Python

Text Mining with Machine Learning and Python [Video], Published by Packt

sunyong2016/The-subsidy-scheme-of-DiDi

摘要：“打车难”问题是国民关注的热点问题之一。随着“互联网+”时代的到来，各种打车软件应运而生，并推出了多种出租车补贴方案，各种补贴方案是否能缓解打车难，仍有待研究。为此我们搜集相关数据，利用数学建模的方法，研究了“互联网+”时代的出租车资源配置问题，获得了一些有价值的结论。我们的主要工作如下： 由于题目未提供数据，通过查找资料，选择2011年同济大学数学建模夏令营D题提供的深圳市出租车GPS数据进行研究。因数据过于庞大且存在错误，故先要剔除空间、状态上不合理的数据，然后按照时间、经纬度、行驶状态进行归类整理。 问题一要求建立出租车资源供求匹配程度模型。我们选择出租车总量供需比 、出租车拥有量（万人） 、出租车空驶率 三项指标对深圳市出租车匹配程度进行研究。首先，基于出租车总量供需比研究出租车的供求匹配程度，通过查找资料得到出租车需求量的测定模型，代入数据得到2011年深圳市出租车需求量为26110辆，与实际出租车供给量15035辆作比，求得出租车总量供需比 为0.5758，表明2011年深圳市出租车总量的供需匹配程度较差。然后，基于出租车拥有量（万人）研究不同空间出租车的供求匹配程度，数据中深圳市各地区载客点数目比重可评估各地区出租车供给量，与该地区常住人口数（万人）求比值，得到深圳市不同空间出租车拥有量（万人），其中罗湖区、福田区和南山区的出租车拥有量大，出租车供求匹配程度高，而坪山新区和光明新区的出租车拥有量小，出租车供求匹配程度低。最后，基于出租车空驶率研究不同时间出租车资源的供求匹配程度。将一天分为24个时间单位，由数据得到各时间段的空驶时间和运营时间，两者相比得到各时间段的小时平均空驶率，其中在23:00—7:00出租车空驶率最高，出租车供求匹配程度较低，而在8:00—10:00和14:00—18:00空驶率较低，出租车供求匹配程度较高。 问题二通过建立补贴方案吸引力模型，对“缓解打车难”的帮助程度进行评估。我们认为吸引力越大，软件使用越广泛，对“缓解打车难”帮助越大。首先，以每笔补贴金额为指标，利用模糊数学中隶属函数建立补贴金额与吸引力隶属关系式，其中吸引力因子为6.414。结合各公司出租车补贴方案，求得当单笔补贴方案为10元和2元时吸引力分别为0.91220，0.0927。由于补贴方案随时间的推移，其吸引力会发生变化，构建修正模型，得到补贴方案的综合吸引力模型，求得在每笔补贴方案为10元和2元时其值分别为：6.4880,0.6595。 问题三要求设计打车软件补贴方案，并评价方案的合理性。其实质是构建评价补贴方案合理性模型。该模型主要由公司经营成本，顾客满意度和吸引力三因素决定。顾客满意度受到补贴金额影响，因为当补贴金额较高时，司机将会拒载沿边扬招乘客而使其满意度下降。利用非线性规划求解该模型，得到每笔补贴6.53元时结果最优。对于偏远地区、拥堵路段的打车难问题，本文通过建立司机满意度模型并与公司成本的实际相结合，得到对单个出租车的月补贴为860—1035元较合适。 关键词：供求匹配程度 补贴方案吸引力 满意度 “打车难”

sunyong2016/Tile-fetcher

A python script to download map tiles from Open Street Map server