使用TensorFlow和Keras的神经网络从MNIST数据集中分类手写数字，实现97%的准确性。
A neural network with TensorFlow and Keras to classify hand-written digits from the MNIST dataset, achieving 97% accuracy. (2024-04-02, Jupyter Notebook, 0KB, 下载0次)
耦合a1
coupled a1 (2022-04-28, Python, 0KB, 下载0次)
本项目基于Masure等人的工作([https: github.com LoicMasure ASCADv2Code- Analysis)对ASCADv2-Extracted数据集进行了泄露检测和高阶相关能量分析。](https: github.com LoicMasure ASCADv2Code- Analysis%EF%BC%89%E5%AF%B9ASCADv2-Extracted%E6%95%B0%E6%8D%AE%E9%9B%86%E8%BF%9B%E8%A1%8C%E4%BA%86%E6%B3%84%E9%9C%B2%E6%A3%80%E6%B5%8B%E5%92%8C%E9%AB%98%E9%98%B6%E7%9B%B8%E5%85%B3%E8%83%BD%E9%87%8F%E5%88%86%E6%9E%90%E3%80%82),
Based on the work of Masure et al. ([https: github. com LoicMasure ASCADv2Code - Analysis), this project has carried out leak detection and higher-order correlation energy analysis on the ASCADv2 Extracted dataset.] (https: github.com LoicMasure ASCADV2Code - Analysis% EF% BC% 89% E5% AF% B9ASCADV2-Extracted% E6% 95% B0% E6% 8D% AE% E9% 9B% 86% E8% BF% 9B% E8% A1% 8C% E4% BA% 86% E6% B3% 84% E9% 9C% B2% E6% A3% 80% E6% B5% 8B% E5% 92% 8C% E9% AB% 98% E9% 98% B6% E7% 9B% B 8% E5% 85% B3% E8% 83% BD% E9% 87% 8F% E5% 88% 86% E6% 9E% 90% E3% 80% 82), (2023-09-30, Python, 0KB, 下载0次)
Código-referente ao artigo[https:medium.com@lauangermandi criando uma-camada de cache para-requisi%C3%A7%C3%B5es-...]
Código referente ao artigo https://medium.com/@lauanguermandi/criando-uma-camada-de-cache-para-requisi%C3%A7%C3%B5es-http-com-net-5-refit-e-decorator-com-scrutor-6f0e921ed1c8 (2021-11-14, C#, 0KB, 下载0次)
计算机科学12-A1项目
Computer Science 12 - A1 Project (2017-10-21, Python, 0KB, 下载0次)
AAAI2022 安全AI挑战者计划第八期:Data-Centric Robust Learning on ML Models 0.97+方案之一
AAAI2022 Security AI Challenger Plan Phase VIII: Data Centric Robust Learning on ML Models 0.97+Solution (2022-01-24, Python, 32KB, 下载0次)
大数据A1:Unigram生成器
Big Data A1 : Unigram generator (2016-09-07, Java, 16KB, 下载0次)
使用Go开发《基金数据分析系统》,项目系统框架是基于gin-api-template的基础上做二次开发,gin-api- template源码地址:[https: github.com 52lu gin-api- template,《基金...](https: github.com 52lu gin-api- template%EF%BC%8C%E3%80%8A%E5%9F%BA%E9%87%91%E6%95%B0%E6%8D%AE%E5%88%86%E6%9E%90%E7%B3%BB%E7%BB%9F%E3%80%8B%E8%AE%A1%E5%88%92%E5%AE%8C%E6%88%90%E5%8A%9F%E8%83%BD%EF%BC%8C%E4%B8%BB%E8%A6%81%E6%9C%89)
Use Go to develop the Fund Data Analysis System. The project system framework is based on the gin api template for secondary development, Gin api template Source code address: [https: github.com 52lu gin api template, Fund...] (https: github.com 52lu gin api template% EF% BC% 8C% E3% 80% 8A% E5% 9F% BA% E9% 87% 91% E6% 95% B0% E6% 8D% AE% E5% 88% 86% E6% 9E% 90% E7% B3% BB% E7% BB% 9F% E3% 80% 8B% E8% AE% A1% E5% 88% 92% E5% AE% 8C% E6% 88% 90% E5% 8A% 9F% E8% 83% BD% EF% BC% 8C% E4% B8% BB% E8% A6% 81% E6% 9C% 89) (2021-11-08, GO, 83KB, 下载0次)
利用Python进行数据分析([https: read.douban.com reader ebook 15249337 )源代码整理版](https: read.douban.com reader ebook 15249337 %EF%BC%89%E6%BA%90%E4%BB%A3%E7%A0%81%E6%95%B4%E7%90%86%E7%89%88)
Data analysis using Python ([https: read.douban.com reader ebook 15249337) source code collation) (https: read.douban.com reader ebook 15249337% EF% BC% 89% E6% BA% 90% E4% BB% A3% E7% A0% 81% E6% 95% B4% E7% 90% 86% E7% 89% 88) (2018-12-31, Jupyter Notebook, 35601KB, 下载0次)
计算机科学作弊表A0海报。
A Computer Science Cheat Sheet A0 Poster. (2018-03-05, TeX, 143KB, 下载0次)
剩余注意力网络，剩余注意力网络的一个重要实现。cifar上的最佳acc为10-97.78%。
ResidualAttentionNetwork,A Gluon implement of Residual Attention Network. Best acc on cifar10-97.78%. (2019-06-12, Python, 45468KB, 下载0次)
道路裂缝检测神经网络-，一种使用keras构建的卷积神经网络，用于检测道路裂缝，准确率为97.5%。
Road-Cracks-Detection-Neural-Network-,A convolutional neural network built using keras to detect cracks on road with 97.5% accuracy. (2021-01-07, Jupyter Notebook, 408KB, 下载0次)
gen efficientnet pytorch、预训练efficientnet、efficientnet-Lite、MixNet、MobileNetV3 V2、MNASNet A1和B1、FBNet、单路径NAS
gen-efficientnet-pytorch,Pretrained EfficientNet, EfficientNet-Lite, MixNet, MobileNetV3 V2, MNASNet A1 and B1, FBNet, Single-Path NAS (2022-07-01, Python, 65KB, 下载0次)
97 TensorFlow.js快速入门，第97集-TensorFlow.js快速入门
Episode 97 - TensorFlow.js Quick Start
, (2018-04-02, TypeScript, 17448KB, 下载0次)
德国交通标志分类使用TensorFlow，在这个项目中，我使用Python和TensorFlow对交通标志进行分类。使用的数据集：德国交通标志数据集。T...
In this project, I used Python and TensorFlow to classify traffic signs. Dataset used: German Traffic Sign Dataset. This dataset has more than 50,000 images of 43 classes. I was able to reach a +99% validation accuracy, and a 97.3% testing accuracy. (2019-12-03, HTML, 8970KB, 下载0次)
这个notebook是根据去年o2o比赛冠军的代码（[https://github.com/wepe/O2O-Coupon-Usage-
Forecast）整理注释的，使用notebook（安装了插件方便整理）,加了一些自己的注释，调了bug，实测可以跑通。用xgboost应该在0.795左右。](https://github.com/wepe/O2O-Coupon-
Usage-
Forecast%EF%BC%89%E6%95%B4%E7%90%86%E6%B3%A8%E9%87%8A%E7%9A%84%EF%BC%8C%E4%BD%BF%E7%94%A8notebook%EF%BC%88%E5%AE%89%E8%A3%85%E4%BA%86%E6%8F%92%E4%BB%B6%E6%96%B9%E4%BE%BF%E6%95%B4%E7%90%86%EF%BC%89,%E5%8A%A0%E4%BA%86%E4%B8%80%E4%BA%9B%E8%87%AA%E5%B7%B1%E7%9A%84%E6%B3%A8%E9%87%8A%EF%BC%8C%E8%B0%83%E4%BA%86bug%EF%BC%8C%E5%AE%9E%E6%B5%8B%E5%8F%AF%E4%BB%A5%E8%B7%91%E9%80%9A%E3%80%82%E7%94%A8xgboost%E5%BA%94%E8%AF%A5%E5%9C%A80.795%E5%B7%A6%E5%8F%B3%E3%80%82)
,
This notebook was compiled based on the code from last year s O2O competition champion (https: github. com wepe O2O Coupon Usage - Forecast). It was annotated using the notebook (with plugins installed for ease of compilation), added some own comments, fixed bugs, and tested to run smoothly. Using xgboost should be around 0.795 (https: github.com wepe O2O-Coupon- Usage- Forecast%EF%BC%89%E6%95%B4%E7%90%86%E6%B3%A8%E9%87%8A%E7%9A%84%EF%BC%8C%E4%BD%BF%E7%94%A8notebook%EF%BC%88%E5%AE%89%E8%A3%85%E4%BA%86%E6%8F%92%E4%BB%B6%E6%96%B9%E4%BE%BF%E6%95%B4%E7%90%86%EF%BC%89,%E5%8A%A0%E4%BA%86%E4%B8%80%E4%BA%9B%E8%87%AA%E5%B7%B1%E7%9A%84%E6%B3%A8%E9%87%8A%EF%BC%8C%E8%B0%83%E4%BA%86bug%EF%BC%8C%E5%AE%9E%E6%B5%8B%E5%8F%AF%E4%BB%A5%E 8%B7%91%E9%80%9A%E3%80%82%E7%94%A8xgboost%E5%BA%94%E8%AF%A5%E5%9C%A80.795%E5%B7%A6%E5%8F%B3%E3%80%82) (2018-08-09, Jupyter Notebook, 15KB, 下载0次)
BP神经网络模型及其在sklearn的iris数据集上的训练结果（样本太少，导致训练准确率只有97%）
,
BP_ NN, BP neural network models and their training results on the iris dataset of sklearn (with too few samples, resulting in a training accuracy of only 97%), (2018-05-21, Python, 12KB, 下载0次)
A1关于网络抓取Pandas数据科学，在python中使用Pandas学习网络抓取。-数据科学，
Learning WebScraping using Pandas in python. - Data Science
, (2021-12-02, Jupyter Notebook, 795KB, 下载0次)
（1）写一个final参数类M，包括比较次数、交换次数、探测次数属性，并重写构造器和toString方法。
（2）写一个抽象类A，其中包括要排序的数据。提供三个final方法，分别完成比较、探测、交换操作的同时，正确改变私有的M类对象成员的相关属性。并提供一个虚方法doSort,同时提供一个final方法sort（先设置M对象初值，然后调用doSort方法，返回M对象引用）
（3）写三个采用不同方法排序的A类的派生类A1,A2,A3
（4）写一个测试类作为主类，分别生成A1,A2,A3的对象并调用sort方法，显示三个方法在排序时候的性能参数。
（5）谈谈这种类设计的合理性以及可以改进之处。
（6） 给前面的排序测试算法增加排序可视化功能，提示：初始化排序类的时候用图形显示排序的数据，进行移动或者比较等操作时，用图形演示动作影响的操作。（请思考如何防止演示过快）。
(1) write a final parameters of type M, including the number of comparisons, the exchange frequency, detecting the number of attributes, and override the constructor and toString methods.
(2) to write an abstract class A, which would include the sort of data. Provides three final methods of the respective comparison, detection, while the exchange operation, the right to change the private member of M-class object associated attributes. And to provide a virtual method doSort, while providing a final method sort (first set the initial value of M objects, then call doSort method returns an object reference M)
(3) three different ways to write sort of A class derived class A1, A2, A3
(4) Write a test class as main class, were generated A1, A2, A3 and call the sort method of the object, indicating when the three methods in order of performance parameters.
(5) to talk about this kind of design is reasonable and can be improved.
(6) to increase the previous sorting algorithm sorting (2010-06-11, Java, 457KB, 下载107次)
Bayes滤波器算法,c++语言实现,Visual Studio.net下实现,2003年8月修正版本,用于数据融合算法
Bayesian filter algorithm, c language, was officially realizing, in August 2003 amended version, algorithm for data integration (2004-07-15, C++, 127KB, 下载154次)