协同过滤推荐算法

结合HanLP提供的针对机构名的HMM语料，实现了基于HMM模型的机构名实体识别算法。(Implementing Code Extraction of Institution Names with HMM)

说明：  用于新冠肺炎胸片扫描，来对新冠肺炎与普通肺炎进行区分(Used for chest scan of New Coronary Pneumonia to distinguish New Coronary Pneumonia from Common Pneumonia)

说明：  该资源为应用GNUradio工具，对USRP的开发，主要实现了频谱感知，以及根据频谱感知的结果进行ofdm通信的功能(This resource is the development of USRP using gnuradio tool. It mainly realizes the function of spectrum sensing and OFDM communication according to the result of spectrum sensing)

大规模对抗代码，有追逃，有需要的可以看看(Python code about Against. You can download if you need.)

基于python语言的人脸识别模块功能的实现和实时摄像头的人脸识别(Realization of face recognition module function based on Python language and face recognition of real-time camera)

用Python完成了github的oauth2功能(The OAuth 2 function of GitHub is completed with Python)

聊天机器人 原理： 严谨的说叫 ”基于深度学习的开放域生成对话模型“，框架为Keras（Tensorflow的高层包装），方案为主流的RNN（循环神经网络）的变种LSTM（长短期记忆网络）+seq2seq（序列到序列模型），外加算法Attention Mechanism（注意力机制），分词工具为jieba，UI为Tkinter，基于”青云“语料（10万+闲聊对话）训练。 运行环境：python3.6以上，Tensorflow，pandas，numpy，jieba。(Chat Robot Principle: Strictly speaking, it is called "Open Domain Generation Dialogue Model Based on Deep Learning". The framework is Keras (High-level Packaging of Tensorflow). The scheme is LSTM (Long-term and Short-term Memory Network)+seq2seq (Sequence to Sequence Model), plus Attention Mechanism (Attention Mechanism). The word segmentation tool is Jieba and the UI is Tkinter. Based on "Qingyun" corpus (100,000 + chat dialogue) training. Running environment: Python 3.6 or more, Tensorflow, pandas, numpy, jieba.)

说明：  使用一维卷积神经网络处理序列数据，数据类型为一维(One dimensional convolution processing sequence data)

利用深度学习进行遥感图像场景分类 这里我们对NWPU-RESISC45数据集的场景图像进行分类 我们将卷积神经网络应用于图像分类。我们从头开始训练数据集。此外，还应用了预先训练的VGG16 abd ResNet50进行迁移学习。(Scene Classification of Remote Sensing Images Using Deep Learning Here we classify scene images from NWPU-RESISC45 dataset We apply convolutional neural network to image classification. We start training data sets from scratch. In addition, a pre-trained VGG16 abd ResNet50 is used for migration learning.)

利用python编程应用在社交网络算法在金融反欺诈中，在社区内使用最短路径算法来发现失联用户与一个正常还款用户的关系链 分析指标：degree、closeness centrality、betweenness centrality、cluster coefficient、triangle count、connected components算法：PageRank社区发现：GN、FastUnfolding、LPA、SLPA、WalkTrap