Week 8 and 9 Literate programming
(Communicating Results)1 Default setting of the Quarto Manuscript
1.1 Create the folder
(1) Create an empty folder for Quarto Manuscript and give it a proper name.
(2) Click File-New File-Quarto project-Manuscript project.
(3) Choose the folder we created in step 1.
1.2 Check the main YAML
(1) title:
(2) authors:
(3) bibliography: article citation
1.3 Check the structure of working folder
(1) _quarto.yml: format YAML. This is related to metadata and front matter of the
manuscript. Different formats of output are customized in this ffle.
(2) index.qmd: we write in this main document.
(3) references.bib: your references (which can be downloaded through Google Scholar or
through Zotero).
(4) Rename index.qmd as 0_main_ffle.qmd.
11.4 Generate two additional markdown ffles
Those additional markdown ffles conduct analysis that complement this main ffle.
1_generate_number: remember to put your affffliation here.
2_generate_ffgure: remember to put your affffliation here.
2 Edit YAML section and YAML ffle
2.1 Change the main ffle name in YAML
Please go to _quarto.yml to change the default ffle name to manuscript: article:
0_main_ffle.qmd.
2.2 Python engine
jupyter: python3
2.3 Affffliation
title: Assignment 3
affffliation: University of Auckland
email: xxxx
2.4 Date
date: last-modiffed
2.5 Abstract
abstract: Please put your absract of assignment 3 here. Example: This course will teach
you data science workffow. The course will focus on Python for Finance and reproducible
analytical pipeline.
22.6 keywords
Please put your keywords here
keywords: - Quarto - Manuscript
2.7 Add PDF as one of the output
Go to _quarto.yml
add pdf: default at the next level paramater of format:
comment out jats: default
3 Render the manuscript
3.1 First try
Click the preview button in Vscode.
Additional folders generated: .quarto, _manuscript, _freeze
_freeze: code is not re-evaluated unless their source changes. A corresponding _freeze folder
will be created.
3.2 Click on Word and PDF icon form your folder
Please go to the folder you created at step 1 of Section 1.1
This probably does not work.
Solution 1: Click on Word and PDF icon within Vscode.
Solution 2: manually edit the format parameter within YAML.
3.3 Add number for each section
number-sections: true
34 Python code cells
4.1 Article Notebook: wrong way
If you only run the following code, The article notebook will not appear..
Mylist = [1,2,3,4]
4.2 Article Notebook: correct way
We have to print something or return an object to make Article Notebook show up.
[1, 2, 3, 4]
4.3 Echo code or not
In the code, we can add the following parameter to control whether code will be echoed in the
main text (not Article Notebook):
#| echo: true
More about code cells on the following page: https://quarto.org/docs/reference/cells/cellsjupyter.html
4.4
Output results or not
#| output: false
Mylist = [1,2,3,4]
Mylist
4.5 Other useful code cells parametrs
#| code-fold: true (Collapse code into an HTML tag)
#| code-line-numbers: true (Include line numbers in code block output)
45 Numbers, ffgures, and tables
5.1 Inline code for numbers
The average gap of eruption is 2.5.
5.2 External embeds: a ffgure
Let us import a ffgure created in another QMD ffle.
Figure 1: Timeline of recent earthquakes on La Palma
Embed: please refer to https://quarto.org/docs/authoring/notebook-embed.html for details
of how to use embeds in Quarto.
5.3 External embeds: speciffc folded line of code
Use the following code to import pandas:
import pandas as pd
import numpy as np
5.4 Display tables
An example adapted from Quarto documentation. Table 1 shows information related to different
car models.
Table 1: Example
mpg cylinders displacement horsepower weight
18 8 307 130 3504
15 8 350 165 3693
5Table 1: Example
mpg cylinders displacement horsepower weight
18 8 318 150 3436
16 8 304 150 3433
17 8 302 140 3449
15 8 429 198 4341
14 8 454 220 4354
14 8 440 215 4312
14 8 455 225 4425
15 8 390 190 3850
5.5 Add existing figures from a folder
Figure 2: A logo for BUSFIN 711
5.6 Cross References
We can cross-reference figures, tables, equations, sections, etc.
5.7 External embeds: several line of code
This is mainly for the purpose of demonstration:
6 1_generate_number
A (University of Auckland)
import pandas as pd
import numpy as np
相关推荐
Qt QChart实战:从零打造一个实时温度监控仪表盘(附完整源码)
本文详细介绍了如何使用Qt QChart模块从零构建一个工业级实时温度监控仪表盘。通过实战案例,展示了波形图的绘制、坐标轴配置、数据流模拟等关键技术,并提供了完整的源码实现,帮助开发者快速掌握Qt在数据可视化领域的应用。
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。
pytorch实现手语识别
连续手语识别项目,中文数据集CSL
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。
别再死磕U-Net了!用PyTorch手把手教你搭建ResUnet,搞定医学图像分割(附完整代码)
本文详细介绍了如何利用PyTorch搭建ResUnet模型,通过引入残差连接解决医学图像分割中的梯度消失、特征退化和细节丢失问题。与传统U-Net相比,ResUnet在Dice系数和收敛速度上均有显著提升,特别适合处理脑部MRI等复杂医学图像。文章还提供了完整的代码实现和关键训练技巧,帮助开发者快速掌握这一先进技术。
通义千问AI模型对接飞书机器人-集成飞书机器人(2-2)
http请求地址上一篇百炼平台配置的应用地址。
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。
从零到一:手把手教你用Verilog在FPGA上构建自己的FFT处理器(不依赖IP核)
本文详细介绍了如何从零开始使用Verilog在FPGA上构建256点FFT处理器,不依赖商业IP核。内容涵盖FFT算法基础、蝶形运算单元设计、整体架构实现、时序收敛优化策略以及测试验证方法,重点提升FPGA上算法部署的运行效率与资源利用率。
【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。
串行RapidIO(Serial RapidIO,SRIO):协议介绍
目录一、RapidIO背景介绍二、RapidIO协议概述2.1 包与控制符号2.2 包格式2.4 消息传递2.5 全局共享存储器2.6 流量控制2.7 串行物理层三、I/O逻辑操作与包格式四、维护操作与包格式五、消息操作与包格式六、总结 本文是基于Xilinx平台SRIO介绍(一)RapidIO协议介绍的内容作读书笔记。 一、RapidIO背景介绍 RapidIO采用三层分级体系结构,包括逻辑层、传输层和物理层该分级结的如下图所示: ● 逻辑层:位于最高层,定义全部协议和包的格式,它们为端点器件发起和完
基于计算机视觉的无人便利店技术:从算法到部署的系统化实战指南
摘要:本文系统介绍了基于计算机视觉的无人便利店技术方案,提出"商品检测-跟踪-动作识别-防损-结算"五段式流水线架构。详细阐述了YOLOv8-Seg商品检测、OC-Sort多目标跟踪、3D-CNN+ST-GCN动作识别等核心算法实现,并完成NVIDIA Jetson Xavier NX边缘端的优化部署。通过TensorRT量化、DeepStream加速等技术手段,系统在保持94.7%动作识别准确率的同时,将漏单率降至0.8%,单店云端算力成本较行业平均降低84%。所有代码已开源,可直接应
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。
你一直忽略的神器:离你最近的超强语音输入法(微信语音输入全攻略,支持全局混用)
你一直忽略的语音输入工具🎤 ,其实就在身边!无需额外软件,即可实现全局语音转文字⚡ 支持中英文混输,还能与其他输入法混合使用,聊天、办公都能高效完成,让你的输入效率轻松翻倍🚀
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。
jre9下载 jdk9
(Java SE Runtime Environment 9.0) 9.0u175 官方最新版 【x86|x64】
C# WinForm 基于DirectShow.dll编写的程序,可边预览边保存视频,音频和图片
基于微软的DirectShow.dll的二次开发1.自动识别连接的摄像头跟麦克风且可以对其进行选择(默认选中第一个)2.可录制wmv和avi两种格式的视频并且都带有音频,avi可以录制的分辨率和帧数进行选择,wmv默认进行了淸晰度优化如不喜欢可去除代码中有相关提示3.可在预览点开之后进行拍照,也可边录制视频边拍照4.添加了视频和音频的压缩功能(酌情使用)这次二次开发发现C#对DirectShow的相关信息太少,这次做完也不敢藏私共享出来,还望大家指正
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(
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');```[](https://bitdeli.com/free "Bitdeli Badge")
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。
霍尼韦尔SoftMaster使用说明
霍尼韦尔SoftMaster使用说明
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。
【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。
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。
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。
【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。
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
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
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,
深度学习基本原理:从数学基础到神经网络训练
深度学习训练的核心,是先确定模型结构,再通过前向计算得到损失,利用链式法则和反向传播计算梯度,最后借助优化器与学习率不断更新参数。真正理解这条链路,后续学习更复杂的网络结构才有可靠的基础。
计算机毕业设计选题推荐:【基于深度学习的车辆损坏严重程度评估方法研究】ResNet50 图像分类
系统用 ResNet50 完成轻微、中度、严重三档分类,输出类别、置信度和单次耗时;视频按每秒一帧抽帧识别,历史区同步出图,没识别出来的帧也保留原图;识别记录可检索、看详情和删除,大屏按类别、时段和任务类型做统计。训练采用迁移学习,部署走 ONNX,推理前加纯色和域外两道拒识,尽量避免把非车损画面硬判成某一类。
246



被折叠的 条评论
为什么被折叠?



