ENG4200 Neural networks

AI对战五子棋——深度学习策略网络 在本篇博文中,我们完成了:✔ 构建一个 CNN 五子棋策略网络✔ 让神经网络输出落子概率✔ 使用 Policy Gradient 强化学习训练 AI✔ 让 AI 实现人与 AI、AI 自对弈博弈。 阅读详情

Coursework 2: Neural networks 

ENG4200 Introduction to Artificial Intelligence and Machine Learning 4 

1. Key Information 

• Worth 30% of overall grade 

• Submission 1 (/2): Report submission 

• Deadline uploaded on Moodle 

• Submission 2 (/2): Code submission to CodeGrade 

• Deadline uploaded on Moodle (the same as for report) 

2. Training data 

The training dataset has been generated by maximum flow analysis between nodes 12 and 2. The 

feature dataset has 19 fields, which of each represents the maximum flow capacity of each of the 

19 edges, taking the values of 0, 1, and 2. The output dataset has 20 fields, where the first 19 

fields refer to the actual flow taking place on each of the 19 edges, and the last one refers to the 

maximum flow possible between nodes 12 and 2. 

Figure 1 The network used to generate training dataset. This information is just to help you understand the training 

dataset; you must not generate additional training dataset to train your neural network. 

 3. What you will do 

You have to create and train a neural network with the following requirement/note: 

• Only the provided training dataset should be used, i.e. furthur traning dataset must NOT be 

created by performing maximum flow analysis over the network in Figure 1. 

• The accuracy on a hidden test dataset will be evaluated by a customised function as 

follows, where the accuracy on the maximum flow field is weighted by 50%, and other 19 

fields share the rest 50% (you may design your loss function accordingly): 

 You should prepare two submissions, code submission and report submission. In blue colour are 

assessment criteria. 

• Code submission should include two files (example code uploaded on Moodle): 

o A .py file with two functions 

▪ demo_train demonstrates the training process for a few epochs. It has three 

inputs: (1) the file name of taining feature data (.csv), (2) the file name of 

training output data (.csv), and (3) the number of epochs. It needs to do two 

things: (1) it needs to print out a graph with two curves of training and 

validation accuracy, respectively; and (2) save the model as .keras file. 

▪ predict_in_df makes predictions on a provided feature data. It has two 

inputs: (1) the file name of a trained NN model (.keras) and (2) the file name 

of the feature data (.csv). It needs to return the predictions by the NN model 

as a dataframe that has the same format as ‘train_Y’. 

o A .keras file of your trained model 

▪ This will be used to test the hidden test dataset on CodeGrade. 

o You can test your files on CodeGrade. There is no limit in the number of 

submissions on CodeGrade until the deadline. 

o Assessment criteria 

▪ 5% for the code running properly addressing all requirements. 

▪ 10% for a third of the highest accuracy, 7% (out of 10%) for a third of the 

second highest accuracy, and 5% (out of 10%) for the rest. 

• Report submission should be at maximum 2 pages on the following three questions and 

one optional question: 

o Parametric studies of hyperparameters (e.g. structure, activators, optimiser, learning 

rate, etc.): how did you test different values, what insights have you obtained, and 

how did you decide the final setting of your model? 

o How did you address overfitting and imbalanced datasets? 

o How did you decide your loss function? 

o [Optional] Any other aspects you’d like to highlight (e.g. using advanced methods 

such as graphical neural network and/or transformer)? 

o [Formatting] Neither cover page nor content list is required. Use a plain word 

document with your name and student ID in the first line. 

o Assessment criteria 

▪ 5% for each of the questions, evaluated by technical quality AND 

writing/presentation 

▪ Any brave attempts of methods (e.g. Graphical Neural Network, Transformer, 

or Physics-Informed Neural Network using physical relationships e.g. that 

the flows going in and out of a node should be balanced) that go beyond 

what we learned in classroom will earn not only the top marks for report, but 

also (unless the accuracy is terribly off) will earn a full 10% mark for 

accuracy in the code submission part. If you have made such attempts, don’t 

forget to highlight your efforts on the report. 

MATLAB 论文复现——面向电量-调频-容量市场的数据中心园区算力及电力资源规划 如何在保证数据中心服务质量的前提下,考虑数据中心负荷转移特性,并对其进行合理规划,对降低数据中心规划总成本及提升新型电力系统灵活性均有显著意义。然后,根据不同电力市场的结算规则,研究了数据中心园区参与电量-调频-容量市场的方式和经济收益。在此基础上,以综合成本最小化为目标,优化数据中心园区发电机、储能的建设容量以及服务器的建设数量。最后,以阿里巴巴公开运行数据为例,证明考虑多元电力市场能够大大降低数据中心规划成本,激励数据中心园区投资建设更多灵活性资源,为电网提供有效的调频服务和容量支撑。 阅读详情

相关推荐

卷积神经网络中的边缘检测:从Canny算法到深度学习(PyTorch实现)

本文系统梳理了边缘检测从经典Canny算法到深度学习方法的演进,重点解析了Canny算法的数学原理及其对卷积神经网络设计的深远影响。通过PyTorch实战,详细演示了如何构建并训练一个基于HED架构的深度学习边缘检测模型,并探讨了模型优化、评估及与传统方法融合的高级技巧。

jenkins8butler的博客 154

计算机科学期刊国外,科学网—影响因子最大的国外计算机科学期刊50种(转) - 孟海战的博文...

Rank Abbreviated Journal Title(linked to full journal information) ISSN ImpactFactor1 BIOINFORMATICS 1367-4803 4.6152 J MACH LEARN RES 1532-4435 3.8183 IBM J RES DEV 0018-8646 3.7004 IEEE NETWORK...

weixin_42467374的博客 327

深入理解搜索引擎——搜索评价指标

    搜索引擎,在做好query理解、索引召回以及排序模型之后,就能直接推上线了吗?答案是否定的,还需对其性能和质量进行评测。性能无非是对时间和空间的运行效率作评测,不细讲,今天讲讲搜索引擎的质量评测。通过质量评测,可根据评测结果不断的进行改进和研究,也可以验证搜索引擎在真实环境中运行时的实际效果。因此,搜索引擎的质量评测对于整个系统的研制和发展是至关重要的。     那么有哪些指标可以评估搜索引擎的结果质量呢?最广为人知的

我是药老的博客 4485

外文版计算机科学期刊,科学网—影响因子最大的国外计算机科学期刊50种(转) - 孟海战的博文...

Rank Abbreviated Journal Title(linked to full journal information) ISSN ImpactFactor1 BIOINFORMATICS 1367-4803 4.6152 J MACH LEARN RES 1532-4435 3.8183 IBM J RES DEV 0018-8646 3.7004 IEEE NETWORK...

weixin_39747577的博客 133

Pspice安装教程:Windows系统完整指南(含环境变量设置)

详细讲解在Windows系统中完成Pspice安装的全过程,涵盖安装步骤与关键的环境变量设置,帮助用户顺利运行仿真工具。适合初学者快速上手pspice安装教程中的核心环节。

weixin_42230607的博客 1181

微信小程序云开发 - 实现商家转账到零钱(微信企业付款到零钱) V3功能流程及完整实例代码,过微信云开发的云函数实现企业付款到零钱(提供详细示例源码纯云函数实现,证书、密钥、签名等各种常见问题解决方案

企业付款到零钱(微信小程序提现功能,提现到零钱)微信小程序-云函数实现微信企业付款到零钱功能,实现微信商户号中的企业付款到零钱的功能,微信小程序+云开发+商家转账至零钱+微信小程序用户提现,使用云开发环境的云函数实现微信小程序的支付功能,微信小程序实现微信企业付款到零钱功能的流程,V3 V2基于微信云开发的商家转账至零钱,小程序云函数企业付款到零钱功能实现,云开发实现微信商户支付企业付款到零钱功能,不用服务器纯云函数实现,详细示例代码demo教程,API证书错误,转账不到账怎么办,密钥生成有误,转账失败,有

🏆 前端领域优质创作者 1万+

RNNoise-SpeexDSP-WebRTC使用说明

本文介绍RNNoise、SpeexDSP与WebRTC音频处理库的下载、调用及参数配置方法,适用于Android语音工程集成。RNNoise基于深度学习,适合非稳态噪声降噪,需48kHz输入;SpeexDSP支持多采样率与回声消除,适用于全双工场景;WebRTC提供轻量级窄带处理。三者可串联使用:麦克风信号先经前级降噪(RNNoise或SpeexDSP),再进入WebRTC媒体引擎进行编码。重点说明各库的典型应用场景、接口调用方式、关键参数调节策略及采样率转换注意事项,帮助开发者高效集成与优化语音处理链路。

知识共享,你我皆是知识的搬运工 239

如何解除“无法完成操作 因为文件包含病毒或潜在垃圾软件”

当运行软件遇到“无法完成操作 因为文件包含病毒或潜在垃圾软件”时,如何解决?控制面板-系统和安全-Windows Defender防火墙-自定义设置,将专用网络及公用网络关闭。点击后,选择“病毒和威胁防护”-管理配置。进入安全中心后,选择“病毒和威胁防护”将需要排斥的选项输入后,保存配置即可。2:打开windows安全中心。

笔者从事电信媒体开发多年,愿意将多年的开发经验分享给同行 6万+

基于深度学习的商品评论情感分析系统-毕业设计

面向商品评论的中文情感分析系统,支持单条预测、批量分析、情感分布与关键词、报告导出。TextCNN、BiLSTM、Transformer 在同一词表、同一数据切分、同一种子与超参下公平对比,测试集 F1 分别为 0.9101、0.9089、0.9075。结论是这一规模下模型复杂度未带来收益,但误差结构不同。附技术栈、截图、文档预览与演示视频。

csbysj2020 158

28.Anaconda安装jupyter notebook、jupyterlab及体验

手贱,不小心将Anaconda自带得jupyter给卸载了,没法子了查了好多资料都比较麻烦,所以自己摸索着重新安装,记录一下心得。说明:Anaconda没有加入环境变量,Anaconda官方建议不要将Anaconda加入环境变量,这就意味着他和python是可以共存的。

迷逝 5031

构建个体结构脑网络的另一种方法

这篇文献提出了一种构建个体结构脑网络的方法即Morphometric INverse Divergence (MIND),通过T1像数据即可对每个被试构建一个脑网络,且通过多个特征进行构建而非单一的特征。即可获得由4个特征得到的个体结构网络矩阵,可以根据自己的需求采用不同的特征组合得到网络矩阵,也可以结合T1/T2像得到髓鞘化相关的网络矩阵,后续可修改代码批量处理自己的数据。这是文献贴出来的代码。在终端直接克隆到本地。2.修改路径运行代码。

weixin_52073820的博客 825

Transformer 为什么能取代 RNN?从自注意力讲起

2017 年谷歌在《Attention Is All You Need》中提出 Transformer,核心是。它如今是 GPT、BERT、ChatGPT 的共同底座。

anew___的博客 979

基于聚类特征工程的水果销量预测:聚类算法对比与多模型评估

水果销量预测研究:聚类分群与多模型对比 摘要 本研究提出了一种融合聚类分析与机器学习的水果销量预测方法。基于包含12个维度的销售数据集,首先采用K-Means、BIRCH和DBSCAN三种聚类算法对水果品类进行分群,通过轮廓系数等指标评估显示BIRCH算法表现最优(0.387)。随后针对不同群组构建差异化特征工程,并比较XGBoost、LightGBM和LSTM三种预测模型性能。实验结果表明:基于聚类特征工程的LightGBM模型预测效果最佳(RMSE=2.34,R²=0.87).

huanghm88的专栏 139

【原创深度学习项目】基于深度学习的建筑工地防护装备检测系统|安全帽与反光衣检测系统(RT-DETR-L 模型)

本文设计并实现了一套基于RT-DETR-L的建筑工地防护装备检测系统,采用Django+Vue前后端架构,实现对安全帽、反光背心等7类装备及“无PPE”“未戴安全帽”等缺失状态的精准识别。系统支持图片、视频、摄像头实时检测,具备参数可调、数据可视化与统计分析功能,有效解决小目标漏检、误检及管理难问题,助力智慧工地安全管理落地。文末可获取源码。

qq_1445749146的博客 232

Statistics and Machine Learning Toolbox

Statistics and Machine Learning Toolbox(带书签),MATLAB统计学习工具

数学建模模型总结

这是本人在两年内准备参加数学建模竞赛搜集到的模型资料,已经经过了本人的筛选和整理,希望能够对准备参加数学建模竞赛的学弟学妹们提供一些帮助!

图解循环神经网络(RNN)

RNN(循环神经网络)就是“带记忆的网络”:让网络拥有一块“记忆”,在处理当前数据时,能同时参考之前看过的内容;它不一次性看完全部输入,而是像人读书一样,一个字一个字地读,每读一个字就更新一次脑子里的记忆,然后用这份记忆去理解当前这个字。但便签一直在被改写,新信息不断挤占旧信息,传到第 40 步时,“巴黎”的痕迹已经淡得几乎没了。读到“贵”的时候,网络并没有重新去查“好看”这个词——但“好看”的影响已经融在了便签里(那个 1.5)。这三个困难的共同点:答案不在当前这一个字里,而在“前面那串东西”里。

技术管理两手抓,两手都要硬. 424

VC中sqlite的使用

在VC中对sqlite的使用,实现数据库大批数据的快速插入,读取数据库中的数据在ListCtrl中显示

神经网络入门指南:从 CNN、RNN 到 Attention

神经网络的学习,本质是“预测—比较—纠错”的循环;CNN 通过局部连接与权重共享理解图像;RNN 通过隐藏状态传递上下文;LSTM 和 GRU 用门控机制管理记忆;Encoder–Decoder 把理解与生成拆成两个阶段;Attention 则让模型在每一步都能主动寻找最相关的信息。到这里,我们已经走完了从普通神经网络到 Attention 的完整路线。下一篇将正式进入 Transformer,看看它如何摆脱 RNN 的顺序计算,又如何让每个 Token 直接找到与自己相关的信息。

爱敲代码的小黄 396

深度学习工具包 DeepLearningToolbox

深度学习工具包Deprecation notice.-----This toolbox is outdated and no longer maintained.There are much better tools available for deep learning than this toolbox, e.g. [Theano](http://deeplearning.net/software/theano/), [torch](http://torch.ch/) or [tensorflow](http://www.tensorflow.org/)I would suggest you use one of the tools mentioned above rather than use this toolbox.Best, Rasmus.DeepLearnToolbox================A Matlab toolbox for Deep Learning.Deep Learning is a new subfield of machine learning that focuses on learning deep hierarchical models of data.It is inspired by the human brain's apparent deep (layered, hierarchical) architecture.A good overview of the theory of Deep Learning theory is[Learning Deep Architectures for AI](http://www.iro.umontreal.ca/~bengioy/papers/ftml_book.pdf)For a more informal introduction, see the following videos by Geoffrey Hinton and Andrew Ng.* [The Next Generation of Neural Networks](http://www.youtube.com/watch?v=AyzOUbkUf3M) (Hinton, 2007)* [Recent Developments in Deep Learning](http://www.youtube.com/watch?v=VdIURAu1-aU) (Hinton, 2010)* [Unsupervised Feature Learning and Deep Learning](http://www.youtube.com/watch?v=ZmNOAtZIgIk) (Ng, 2011)If you use this toolbox in your research please cite [Prediction as a candidate for learning deep hierarchical models of data](http://www2.imm.dtu.dk/pubdb/views/publication_details.php?id=6284)```@MASTERSTHESIS\{IMM2012-06284, author = "R. B. Palm", title = "Prediction as a candidate for learning deep hierarchical models of data", year = "2012",}```Contact: rasmusbergpalm at gmail dot comDirectories included in the toolbox-----------------------------------`NN/` - A library for Feedforward Backpropagation Neural Networks`CNN/` - A library for Convolutional Neural Networks`DBN/` - A library for Deep Belief Networks`SAE/` - A library for Stacked Auto-Encoders`CAE/` - A library for Convolutional Auto-Encoders`util/` - Utility functions used by the libraries`data/` - Data used by the examples`tests/` - unit tests to verify toolbox is workingFor references on each library check REFS.mdSetup-----1. Download.2. addpath(genpath('DeepLearnToolbox'));Example: Deep Belief Network---------------------```matlabfunction test_example_DBNload mnist_uint8;train_x = double(train_x) / 255;test_x = double(test_x) / 255;train_y = double(train_y);test_y = double(test_y);%% ex1 train a 100 hidden unit RBM and visualize its weightsrand('state',0)dbn.sizes = [100];opts.numepochs = 1;opts.batchsize = 100;opts.momentum = 0;opts.alpha = 1;dbn = dbnsetup(dbn, train_x, opts);dbn = dbntrain(dbn, train_x, opts);figure; visualize(dbn.rbm{1}.W'); % Visualize the RBM weights%% ex2 train a 100-100 hidden unit DBN and use its weights to initialize a NNrand('state',0)%train dbndbn.sizes = [100 100];opts.numepochs = 1;opts.batchsize = 100;opts.momentum = 0;opts.alpha = 1;dbn = dbnsetup(dbn, train_x, opts);dbn = dbntrain(dbn, train_x, opts);%unfold dbn to nnnn = dbnunfoldtonn(dbn, 10);nn.activation_function = 'sigm';%train nnopts.numepochs = 1;opts.batchsize = 100;nn = nntrain(nn, train_x, train_y, opts);[er, bad] = nntest(nn, test_x, test_y);assert(er < 0.10, 'Too big error');```Example: Stacked Auto-Encoders---------------------```matlabfunction test_example_SAEload mnist_uint8;train_x = double(train_x)/255;test_x = double(test_x)/255;train_y = double(train_y);test_y = double(test_y);%% ex1 train a 100 hidden unit SDAE and use it to initialize a FFNN% Setup and train a stacked denoising autoencoder (SDAE)rand('state',0)sae = saesetup([784 100]);sae.ae{1}.activation_function = 'sigm';sae.ae{1}.learningRate = 1;sae.ae{1}.inputZeroMaskedFraction = 0.5;opts.numepochs = 1;opts.batchsize = 100;sae = saetrain(sae, train_x, opts);visualize(sae.ae{1}.W{1}(:,2:end)')% Use the SDAE to initialize a FFNNnn = nnsetup([784 100 10]);nn.activation_function = 'sigm';nn.learningRate = 1;nn.W{1} = sae.ae{1}.W{1};% Train the FFNNopts.numepochs = 1;opts.batchsize = 100;nn = nntrain(nn, train_x, train_y, opts);[er, bad] = nntest(nn, test_x, test_y);assert(er < 0.16, 'Too big error');```Example: Convolutional Neural Nets---------------------```matlabfunction test_example_CNNload mnist_uint8;train_x = double(reshape(train_x',28,28,60000))/255;test_x = double(reshape(test_x',28,28,10000))/255;train_y = double(train_y');test_y = double(test_y');%% ex1 Train a 6c-2s-12c-2s Convolutional neural network %will run 1 epoch in about 200 second and get around 11% error. %With 100 epochs you'll get around 1.2% errorrand('state',0)cnn.layers = { struct('type', 'i') %input layer struct('type', 'c', 'outputmaps', 6, 'kernelsize', 5) %convolution layer struct('type', 's', 'scale', 2) %sub sampling layer struct('type', 'c', 'outputmaps', 12, 'kernelsize', 5) %convolution layer struct('type', 's', 'scale', 2) %subsampling layer};cnn = cnnsetup(cnn, train_x, train_y);opts.alpha = 1;opts.batchsize = 50;opts.numepochs = 1;cnn = cnntrain(cnn, train_x, train_y, opts);[er, bad] = cnntest(cnn, test_x, test_y);%plot mean squared errorfigure; plot(cnn.rL);assert(er<0.12, 'Too big error');```Example: Neural Networks---------------------```matlabfunction test_example_NNload mnist_uint8;train_x = double(train_x) / 255;test_x = double(test_x) / 255;train_y = double(train_y);test_y = double(test_y);% normalize[train_x, mu, sigma] = zscore(train_x);test_x = normalize(test_x, mu, sigma);%% ex1 vanilla neural netrand('state',0)nn = nnsetup([784 100 10]);opts.numepochs = 1; % Number of full sweeps through dataopts.batchsize = 100; % Take a mean gradient step over this many samples[nn, L] = nntrain(nn, train_x, train_y, opts);[er, bad] = nntest(nn, test_x, test_y);assert(er < 0.08, 'Too big error');%% ex2 neural net with L2 weight decayrand('state',0)nn = nnsetup([784 100 10]);nn.weightPenaltyL2 = 1e-4; % L2 weight decayopts.numepochs = 1; % Number of full sweeps through dataopts.batchsize = 100; % Take a mean gradient step over this many samplesnn = nntrain(nn, train_x, train_y, opts);[er, bad] = nntest(nn, test_x, test_y);assert(er < 0.1, 'Too big error');%% ex3 neural net with dropoutrand('state',0)nn = nnsetup([784 100 10]);nn.dropoutFraction = 0.5; % Dropout fraction opts.numepochs = 1; % Number of full sweeps through dataopts.batchsize = 100; % Take a mean gradient step over this many samplesnn = nntrain(nn, train_x, train_y, opts);[er, bad] = nntest(nn, test_x, test_y);assert(er < 0.1, 'Too big error');%% ex4 neural net with sigmoid activation functionrand('state',0)nn = nnsetup([784 100 10]);nn.activation_function = 'sigm'; % Sigmoid activation functionnn.learningRate = 1; % Sigm require a lower learning rateopts.numepochs = 1; % Number of full sweeps through dataopts.batchsize = 100; % Take a mean gradient step over this many samplesnn = nntrain(nn, train_x, train_y, opts);[er, bad] = nntest(nn, test_x, test_y);assert(er < 0.1, 'Too big error');%% ex5 plotting functionalityrand('state',0)nn = nnsetup([784 20 10]);opts.numepochs = 5; % Number of full sweeps through datann.output = 'softmax'; % use softmax outputopts.batchsize = 1000; % Take a mean gradient step over this many samplesopts.plot = 1; % enable plottingnn = nntrain(nn, train_x, train_y, opts);[er, bad] = nntest(nn, test_x, test_y);assert(er < 0.1, 'Too big error');%% ex6 neural net with sigmoid activation and plotting of validation and training error% split training data into training and validation datavx = train_x(1:10000,:);tx = train_x(10001:end,:);vy = train_y(1:10000,:);ty = train_y(10001:end,:);rand('state',0)nn = nnsetup([784 20 10]); nn.output = 'softmax'; % use softmax outputopts.numepochs = 5; % Number of full sweeps through dataopts.batchsize = 1000; % Take a mean gradient step over this many samplesopts.plot = 1; % enable plottingnn = nntrain(nn, tx, ty, opts, vx, vy); % nntrain takes validation set as last two arguments (optionally)[er, bad] = nntest(nn, test_x, test_y);assert(er < 0.1, 'Too big error');```[![Bitdeli Badge](https://d2weczhvl823v0.cloudfront.net/rasmusbergpalm/deeplearntoolbox/trend.png)](https://bitdeli.com/free "Bitdeli Badge")

AI 学习第 15 课:变长序列不 pack 直接喂 RNN,隐状态全被 0 污染

NLP 的 batch 里句子长短不一,短句垫 0 补齐后直接喂 RNN,看似能跑,实际上每个 0 垫步都在更新隐状态——短序列的最后一步输出根本不是它真实句子的结果。这一课用 torch 2.15(

aifannihao的博客 160

CNN 卷积神经网络深度解析:卷积、池化与特征提取的底层原理

卷积核(Kernel),也叫滤波器(Filter),本质上是一个可以训练的权重矩阵。1 0 -11 0 -11 0 -1这个卷积核可以对图像中的局部像素进行加权计算。不同的卷积核能够提取不同类型的特征:水平边缘特征垂直边缘特征纹理特征角点特征局部形状特征在传统图像处理中,滤波器通常由人工设计。而在 CNN 中,卷积核的权重通过训练数据和反向传播自动学习得到。这也是 CNN 能够自动提取图像特征的关键。# 第一层卷积padding=1# 第二层卷积padding=1。

2503_90237586的博客 294

图解常用深度学习优化器

来隐式构造 Hessian 逆的近似,从而直接估计出“谷底在哪”,而非仅仅判断“哪边最陡”。即SGD/Adam 主要看“当前坡度”,LBFGS 还会看“最近几步坡度怎么变”,像用局部地形图来规划路线。训练 Transformer 时,有些层梯度大,有些层梯度小。如果用 Adam + L2,梯度大的参数反而可能被“少衰减”,正则效果不稳定;:你要下山,但每次迈步前都要派侦察兵把整座山的每一个坡面都测量一遍,算出“绝对最陡”的方向才走一步。:还是下山,但这次你只问身边一个路人“哪边最陡”,然后就走。

技术管理两手抓,两手都要硬. 219

【无标题】无人机视角道路坑洞坑洼目标检测数据集深度学习 目标检测 无人机航拍 道路坑洞 路面病害 YOLO格式 数据集发布9032期

(一)类别构成该数据集聚焦单一类别——坑洞(类别 ID 为 0),任务指向明确,围绕无人机俯视视角下的道路坑洞目标开展检测标注。类别体系精简,与公路路面病害巡检的实际需求高度契合,图像中的坑洞样本可直接作为路面破损识别的核心正样本,便于对接道路养护排查流程。(二)数据数量数据集规模为 3285 张无人机俯视航拍图像,采用标准 YOLO txt 标注格式,标注框总计 4457 个,其中带有效标注的图片 2007 张,全部图片均带有有效标注。

qq_2529413168的博客 187

基于深度学习的猫图像目标检测方法研究:YOLOv8s模型海选、多种子对比训练与Django+ONNX部署实现详解

本课题基于深度学习研究猫图像目标检测问题。数据集含1000张猫图像、1188个标注框,按固定比例划分并冻结清单;训练前在多个候选族中短训海选,确定YOLOv8s为主模型,经学习率搜索与显存标定后完成150轮训练,验证集mAP50达到0.92,并以7个随机种子的对比实验验证稳定性。最优权重导出为ONNX接入Django系统,提供图片、视频、摄像头三种识别方式,配套识别记录管理与统计大屏,形成从训练到应用的完整闭环。

WEIXIN_JSJBSZDS的博客 220

第 05/24 期:计算图与自动微分:深度学习如何计算梯度

本文系统阐述了神经网络训练中梯度计算的本质:损失对参数的敏感性。通过构建计算图,将复杂函数分解为基本运算,利用链式法则实现反向传播。核心在于反向模式自动微分——以一次前向计算为基础,通过传播输出对中间变量的敏感性,高效求得所有参数梯度。相比前向模式,反向模式在参数多、目标为标量的场景下更高效。计算图不仅适用于神经网络,也广泛适用于任意可微程序。关键思想是:梯度是局部导数与上游敏感性的乘积,分支路径的贡献需累加,从而保证导数计算的正确性与效率。

weixin_47502228的博客 331

【深度学习项目实战·30】PM2.5短期预测:用LSTM构造没有未来信息的滑动窗口

本文基于北京PM2.5数据,使用LSTM实现从最近24小时历史观测预测下一小时PM2.5。重点解决时间切分、因果填补、滑动窗口构建与标准化等关键问题,确保模型仅依赖过去信息,避免未来泄漏。通过滚动单步预测与持续性基线对比,评估在原始单位下的MAE/RMSE,验证模型有效性。强调时间顺序与数据完整性,为时序预测提供可复现的实践框架。

银尘博客 519

无人机巡检道路路面异常检测数据集 深度学习 目标检测 无人机巡检 道路病害 路面异常 YOLO格式 数据集发布9028期

无人机巡检道路路面异常检测数据集 深度学习 目标检测 无人机巡检 道路病害 路面异常 YOLO格式 数据集发布9028期

qq_2529413168的博客 213

深度学习 目标检测 无人机巡检 防震锤 电力设备 YOLO格式 数据集发布

深度学习 目标检测 无人机巡检 防震锤 电力设备 YOLO格式 数据集发布 9038期

qq_2529413168的博客 335

AI 学习第 14 课:CNN 训 6 轮卡在 0.5352,数据没打乱

第 13 课把 Conv2d 的参数量和池化后的尺寸算清了,这课把卷积块拼成一个能训练的完整网络,跑满 6 个 epoch。第一版直接卡死:loss 从 2.6104 掉到 0.6989 就再也不动,

aifannihao的博客 252
上一篇: CS1026A YouTube Emotions
下一篇: ECE 36800 Traffic
wx____codinghelp
博客等级 码龄2年 771粉丝 115原创
评论
成就一亿技术人!
拼手气红包6.0元
还能输入1000个字符
 
 条评论被折叠 查看
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付:10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值